PWR Primary Water Chemistry Guidelines - Miroslav · PDF fileEPRI • 3412 Hillview Avenue,...

PWR Primary Water ChemistryGuidelines

Volume 1, Revision 4

Technical Report

WARNING:Please read the Export ControlAgreement on the back cover.

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 agreement from EPRI. This notice supersedes the export control restrictions and any proprietary licensed material notices embedded in the document prior to publication.

EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 USA800.313.3774 • 650.855.2121 • [email protected] • www.epri.com

PWR Primary Water ChemistryGuidelinesVolume 1, Revision 4 TR-105714-V1R4

Final Report, March 1999

EPRI Project ManagerP. Millett

DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES

THIS PACKAGE WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OFWORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC.(EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) NAMEDBELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:

(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITHRESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEMDISCLOSED IN THIS PACKAGE, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULARPURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNEDRIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS PACKAGE ISSUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR

(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDINGANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISEDOF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THISPACKAGE OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED INTHIS PACKAGE.

ORGANIZATION(S) THAT PREPARED THIS PACKAGE

PWR Primary Water Chemistry Guidelines Committee

ORDERING INFORMATIONRequests for copies of this package should be directed to the EPRI Distribution Center, 207 Coggins Drive, P.O.Box 23205, Pleasant Hill, CA 94523, (925) 934-4212.

Electric Power Research Institute and EPRI are registered service marks of the Electric Power Research Institute, Inc.EPRI. POWERING PROGRESS is a service mark of the Electric Power Research Institute, Inc.

Copyright © 1999 Electric Power Research Institute, Inc. All rights reserved.

iii

CITATIONS

This report was prepared by

PWR Primary Water Chemistry Guidelines Committee

ABB-Combustion EngineeringL. Corsetti, E. Silva

AECLK. Burrill, A. Elliot, D. Guzonas

Ameren UEG. Gary, E. Olson, D. Martin

American Electric PowerT. Andert, G. Decker

Arizona Public ServiceT. Green, J. Santi

Babcock & WilcoxP. Doherty

Baltimore Gas and ElectricR. Conatser, R. Sprecher

Carolina Power and Light J. Eaddy, Jr., R. Hitch

Commonwealth EdisonS. Brant, D. Kozin, J. Petro

Consolidated EdisonR. Burns, T. Teague

Consumers EnergyM. Lee, J. McElrath

Dominion EngineeringJ. Gorman

Duke EnergyA. Boshers, M. Johnson

Duquesne LightV. Linnenbom, D. Orndorf

Electricité de FranceJ-L. Bratelle

Entergy Operations, Inc.B. Burke, G. Dolese,

J. Hathcote, R. Partridge

First EnergyS. Slosnerick

Florida Power Corp.M. Culver, P. Ezzel, J. Payne,

T. Lehmann

FTIM. Bale, M. Bell, L. Lamanna,

J. Willse

GEBCO Engineering, Inc.G. Brobst & J. Riddle

GPU-NG. Bond, R. Bosold, L. Lucas,

P. Walton

INPOR. Bayne, R. Gossman,

R. Schirmer

KEPCOB. Lee, K. Sung

LaborelecL. Duvivier, C. Laire

Lindsay and AssociatesW. Lindsay

NYPAJ. Goldstein, M. Kerns

N. Atlantic Energy Serv. Co.R. Litman

Northeast UtilitiesM. Hudson, C. Matthess,

B. McDonald

Northern States PowerD. Gauger, S. Lappegaard

Nuclear ElectricA. Bates, K. Garbett

NWT Corp.S. Sawochka

Ontario HydroK. Bagli

OPPDB. Schmidt

Pacific Gas & ElectricD. Bradley, J. Gardner,

G. Gillies, E. Wessel

Public Service Electric & GasR. Dolan

Rochester Gas and ElectricB. Dahl, G. Jones, W. Thomson

Siemens Power Corp.J. Partridge, L. Van Swam

South Carolina Electric & GasF. Bacon

Southern California EdisonS. Chick, J. Clark, J. Muniga

Southern NuclearW. Carr, F. Hundley

D. Rickertsen, R. Robinson,J. Sills, K. Turnage

STPNOCD. Bryant, R. Ragsdale

TU ElectricB. Fellers, D. Perkins

TVAD. Adams, S. Harvey, G. Vickery

VattenfallJ.-O. Bengtsson, R. Halldin

Virginia PowerE. Frese, L. Miller, L. Spain

WestinghouseJ. Kormuth

Wisconsin Electric PowerM. Conry, D. Gehrke, D. Gesch,

R. Richards

Wisconsin Public ServiceM. Bernsdorf, W. Flint, G. Holmes,

M. Reinhart, D. Shields

WCNOCM. Blow, T. Jensen, C. Palmer

Consultant to EPRIG. Sabol

ELECTRIC POWER RESEARCH INSTITUTEP. Millett, Chairman; B. Cheng, C. Delatte, P. Frattini, H. Ocken, R. Pathania, N. Torigoe

iv

This report describes research sponsored by EPRI.

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

PWR Primary Water Chemistry Guidelines: Volume 1, Revision 4, EPRI, Palo Alto, CA:1999. TR-105714-V1R4.

v

REPORT SUMMARY

State-of-the art water chemistry programs help ensure the continued integrity of reactorcoolant system (RCS) materials of construction and fuel cladding, ensure satisfactorycore performance, and support the industry trend toward reduced radiation fields.These revised PWR Primary Water Chemistry Guidelines, prepared by a committee ofindustry experts, reflect the recent field and laboratory data on primary coolant systemcorrosion and performance issues. PWR operators can use these Guidelines to updatetheir primary water chemistry programs.

BackgroundPWR water chemistry guidelines are updated periodically as new information becomesavailable and as required by NEI 97-06, Steam Generator Program Guidelines. The lastrevision of these Guidelines identified an optimum primary water chemistry programbased on then-current understanding of research and field information. This revisionprovides further details with regard to primary water stress corrosion cracking(PWSCC), fuel integrity, and on axial offset anomaly (AOA). In addition, because of thecommitment by the industry to NEI 97-06, the revision reflects an increased need forguidance on how to optimize water chemistry programs in a cost consciousenvironment and in a manner consistent with safety and regulatory requirements.

ObjectiveTo update the PWR Primary Water Chemistry Guidelines ñ Revision 3.

ApproachA committee of industry experts, including utility specialists, nuclear steam supplysystem and fuel vendor representatives, Institute of Nuclear Power Operationsrepresentatives, consultants, and EPRI staff, collaborated in reviewing the availabledata on primary water chemistry, reactor water coolant system materials issues, fuelintegrity and performance issues, and radiation dose rate issues. From these data, thecommittee generated water chemistry guidelines that should be adopted at all PWRnuclear plants. Recognizing that each nuclear plant owner has a unique set of design,operating, and corporate concerns, the guidelines committee developed a methodologyfor plant-specific operation.

ResultsRevision 4 of the PWR Primary Water Chemistry Guidelines, which providesrecommendations for PWR primary systems of all manufacture and design, has been

vi

revised as follows:

_ The revised document emphasizes plant-specific optimization of waterchemistry to address individual plant circumstances. It has been revised toreflect the current regulatory environment and the impact of NEI 97-06, whichrequires that U.S. utilities meet the intent of the Guidelines. The revisionattempts to clearly distinguish between prescriptive requirements and non-prescriptive guidance.

_ Sections 2 and 3 have been revised to address the latest information on PWSCC,fuel performance, and radiation field control. Emphasis has been placed onoptimizing pH control programs to minimize the potential for axial offsetanomaly.

_ Section 4 and Appendix C on shutdown and startup chemistry were combinedand moved Volume 2. The combined shutdown and startup chemistry coveragewas expanded to provide increased guidance.

_ A new Section 4 was prepared to provide guidance on developing strategicwater chemistry plans and optimizing water chemistry selections. This sectionalso details the prescriptive requirements of Volume 1 of these guidelines withrespect to NEI 97-06.

_ Several appendices have been updated or added to Volume 1.

EPRI PerspectiveThis fourth revision of the PWR Primary Water Chemistry Guidelines, endorsed by theutility executives of the EPRI Steam Generator Management Project, represents anotherstep in developing a more proactive chemistry program to limit or control steamgenerator degradation, with increased consideration of corporate resources and plantspecific design/operating concerns. It also addresses fuel reliability, core performance,and dose rate control issues in a similar manner. Each utility should carefully examineits plant-specific situation to determine which recommendations should beimplemented.

TR-105714-R4 Vols 1&2

Interest CategoriesChemistrySteam generatorsFuel assembly reliability and performance

KeywordsPWRsWater chemistryPrimary water stress corrosion crackingGuidelinesReactor coolant system

vii

ABSTRACT

Ensuring continued integrity of RCS materials of construction and fuel cladding andmaintaining the industry trend toward reduced radiation fields requires continuedoptimization of reactor coolant chemistry. Optimization of coolant chemistry to meetsite-specific demands becomes increasingly important in light of the movement towardextended fuel cycles and reduced operating costs. This document is the fifth in a seriesof industry guidelines on PWR primary water chemistry. Like each of the others in theseries, it provides a template for development of a plant-specific water chemistryprogram.

ix

EPRI FOREWORD

Chemistry optimization of pressurized water reactor (PWR) primary systems in recenttimes has been complicated by the demands of longer fuel cycles (i.e., higher initialboron concentrations), increased fuel duty (more subcooled boiling) and material/fuelcorrosion concerns. Current utility concerns focus on minimizing costs withoutsacrificing materials integrity or safety. These Guidelines provide a template for aresponsive chemistry program for PWR primary systems and the technicalbases/supporting information for the program. It is the fourth revision of theGuidelines and considers the most recent operating experience and laboratory data. TheGuidelines will be of interest to plant chemists, plant managers, chemical engineers, andengineering managers within utilities owning PWRs.

The Guidelines were prepared by a committee of experienced industry personnelthrough an effort sponsored by EPRI. Participation was obtained from chemistry,materials, steam generator, and fuels experts to ensure the Guidelines present chemistryparameters that are optimum for each set of operating and material conditions. EachEPRI-member utility operating a PWR participated in generation or review of theseGuidelines. Therefore, this document serves as an industry consensus for PWR primarywater chemistry control. In essence, it is a report from industry specialists to theutilities documenting an optimized water chemistry program.

Special acknowledgment is given to the following organizations for submitting first-hand experience through committee participation:

• ABB-Combustion Engineering• AECL• Ameren UE• American Electric Power• Arizona Public Service• Babcock & Wilcox• Baltimore Gas & Electric• Carolina Power & Light• Commonwealth Edison• Consolidated Edison• Consumers Energy• Duke Energy

x

• Duquesne Light• Electricité de France• Entergy Operations, Inc.• First Energy• Florida Power Corporation• Framatome Technologies, Inc.• GPU-Nuclear Corporation• INPO• Korea Electric Power Corporation• Laborelec (Belgium)• New York Power Authority• North Atlantic Energy Service Co.• Northeast Utilities• Northern States Power• Nuclear Electric (UK)• Omaha Public Power District• Ontario Hydro• Pacific Gas and Electric Company• Public Service Electric & Gas• Rochester Gas and Electric• Siemens Power Corp.• South Carolina Electric & Gas• Southern California Edison• Southern Nuclear Operating Co.• South Texas Project Nuclear Operating Company• Tennessee Valley Authority• TU Electric• Vattenfall (Sweden)• Virginia Power• Westinghouse Electric• Wisconsin Electric Power• Wisconsin Public Service Corporation• Wolf Creek Nuclear Operating Corp.

This committee was significantly assisted in its work by J. Gorman of DominionEngineering; G. Brobst and J. Riddle of GEBCO Engineering; W. Lindsay of Lindsayand Associates; K. Garbett and H. Sims of Nuclear Electric/AEA Technology;S. Sawochka of NWT Corporation; G. Sabol, consultant; and B. Cheng, C. Delatte,P. Frattini, H. Ocken, R. Pathania and N. Torigoe of the EPRI staff. In addition, theCommittee would like to acknowledge the support and input of fuel specialists fromEPRI-member utilities.

This document is intended to be a set of guidelines which describe an effective, state-of-the-art program from which a utility can develop an optimized program for their plant.

xi

The philosophy embodied in this document has generic applicability, but can beadapted to the particular conditions of the utility and the site. The detailed guidelinespresented in Sections 3 and 4 and Volume 2 comprise a program that should serve as amodel for the development of site-specific chemistry programs.

Relative to Rev. 3 of these Guidelines, the major changes in this document are as follows:

1. The document has been revised to reflect increased emphasis on plant-specificoptimization of water chemistry to address individual plant circumstances. Thisreorientation was needed to reflect increased diversity among plants and to provideguidance on how to manage primary water chemistry in an increasinglyderegulated and cost conscious environment. The document has also been revisedto reflect the current regulatory environment and the impact of the Nuclear EnergyInstitute (NEI) steam generator initiative, NEI 97-06, which requires that utilitiesmeet the intent of the Guidelines. This revision attempts to clearly distinguishbetween prescriptive requirements and non-prescriptive guidance.

2. Section 2 has been updated to include recent field experiences and relatedinvestigations. Some of the key changes in Section 2 are (1) a more detaileddiscussion of the impact of increased fuel duty (higher burnups, longer fuel cycles,and increased subcooled boiling) on fuel clad corrosion and on water chemistrychoices directed at controlling this corrosion, (2) an expanded discussion of thebenefits of constant high pH regimes for plants with high duty cores where risks offuel clad deposits and associated problems such as axial offset anomaly (AOA) orunder-deposit clad corrosion failures are a concern, (3) an expanded review of theinfluence of lithium concentration and pH on corrosion of fuel cladding, withrecognition that higher lithium has the potential of increasing corrosion of claddingbut, on the other hand, can reduce deposit buildup on fuel rods and thus reducechances of severe corrosion and AOA, (4) an expanded review of the influence ofsilica on corrosion of fuel cladding, with discussion of approaches for operationwith silica above 1000 ppb, (5) a more quantitative discussion of the influence ofwater chemistry on primary water stress corrosion cracking (PWSCC), and (5) anupdated discussion regarding the use of zinc as an additive to ameliorate radiationdose rates and PWSCC.

3. Section 3 has been revised to provide added discussion regarding the use ofcoordinated chemistry at elevated pH (such as constant pH 7.0 or constant pH 7.1,with a lithium limit of 3.5ppm) for high duty cores, while retaining modifiedregimes for lower duty cores for which experience with modified regimes has beensatisfactory. Action Level 1 limits for chlorides, fluorides and sulfates were deletedfrom the table of Power Operation Control Parameters, and a footnote was addedrequiring that utilities develop site-specific administrative limits for these species.The table of Power Operation Diagnostic Parameters was modified to change the"Typical Value" column to read "Reason for Analysis," with changes to values and

xii

footnotes to suit this change. The table covering reactor makeup water was movedfrom Appendix B to this section to reflect the importance of monitoring for hardnesselement impurity ingress via the makeup water.

4. Section 4 and Appendix C on shutdown and startup chemistry were combined andmoved to Volume 2. The combined shutdown and startup chemistry coverage wasexpanded to provide increased guidance: (1) expanded technical discussions wereprovided regarding plant experiences with different types of shutdown and startupchemistry; (2) tables of options, benefits, and possible negative impacts were addedfor shutdown, startup and midcycle outages, together with accompanyingdiscussion and technical support; and (3) a new section was added coveringshutdown with high fission product concentrations in the coolant due to fuelfailures.

5. A new Section 4 was prepared to provide guidance on developing strategic waterchemistry plans and optimizing water chemistry selections. This section also detailsthe prescriptive requirements of Volume 1 of these guidelines with respect to NEI97-06.

6. Appendix A was modified to correct minor errors in the equations and minorinconsistencies between pH tables in the appendix and values calculated using thepH Calculator in EPRI chemWORKS. The thermodynamic data used by the pHcalculator were traced back to original literature references to verify their accuracy,and the output of the pH calculator was checked against independent calculationsemploying the corrected equations in order to validate the codes. A discussion ofsampling was added and some no longer needed portions of the appendix coveringdose rate monitoring were deleted.

7. Appendix B, "Chemistry Control of Supporting Systems," was modified to reflectmoving the table covering makeup water to Section 3.

8. Appendix C in the previous revision, "Startup and Shutdown Chemistry SupportingInformation" was deleted and the information that was contained in it wasintegrated into the old Section 4, "Startup and Shutdown Chemistry ControlRecommendations," which was moved to Volume 2. The remaining appendiceswere relabeled accordingly.

9. Appendix D in the previous revision, "Status of Enriched Boric Acid (EBA)Application," was updated and relabeled Appendix C.

10. Appendix E in the previous revision, "Zinc Injection Experience," was deleted andthe subject of zinc injection was updated and covered in Section 2.

xiii

11. Appendix F in the previous revision, "CVCS Filter and Resin OptimizationPractices," was relabeled Appendix D.

12. A new Appendix E was added, entitled "Oxygen and Hydrogen Behavior in PWRPrimary Circuits." The appendix summarizes the results of a report developed byAEA Technology and Nuclear Electric and made available by them for inclusion inthe Guidelines.

13. A new Appendix F was added, entitled "Considerations for Referencing HighTemperature pH Calculations to 300°C."

Peter J. Millett, ChairmanPWR Primary Water Chemistry Guidelines Revision CommitteeEPRI Nuclear Power GroupMarch 1999

xv

ACRONYMS

AOA Axial offset anomalyBAST Boric acid storage tankBOC Beginning of cycleBRS Boron recovery system.BTRS Boron thermal regeneration systemBWR Boiling water reactorCVCS Chemical and volume control system. The term makeup system (MU) is

used at some plants for the same system.DF Decontamination factorDHC Debye-Huckel limiting slopeDRP Discrete radioactive particleEBA Enriched boric acidECP Electrochemical potentialEPRI Electric Power Research InstituteEOC End of cycleHUT Holdup tank.IGSCC Intergranular stress corrosion crackingINPO Institute of Nuclear Power OperationsNEI Nuclear Energy InstituteNSSS Nuclear steam system supplierNTU Nephelometric turbidity unitsPRT Pressurized relief tankPWR Pressurized water reactorPWSCC Primary water stress corrosion crackingRCP Reactor coolant pumpRCS Reactor coolant system. For purposes of these Guidelines, the RCS

includes the pressurizer.RHR Residual heat removal system. The terms decay heat (DH) system and

shutdown cooling (SDC) system are used at some plants for the same system.

RWST Refueling water storage tank. The term borated water storage tank (BWST) is used at some plants for the same tank.

SCC Stress corrosion crackingSFP Spent fuel pool system. The term spent fuel cooling (SFC) system is

xvi

used at some plants for the same system.STP Standard temperature and pressureTOC Total organic carbonTSS Total suspended solidsVCT Volume control tank. The term makeup tank (MUT) is used at some

plants for the same tank.

xvii

CONTENTS

1 INTRODUCTION AND MANAGEMENT RESPONSIBILITIES ............................................ 1-11.1 Introduction.................................................................................................................. 1-11.2 Generic Management Considerations.......................................................................... 1-21.3 Communication and Training ....................................................................................... 1-31.4 Outage Planning and Coordination.............................................................................. 1-4

2 TECHNICAL BASIS FOR THE NEED TO CONTROL THE COOLANT CHEMISTRYIN PWRS ................................................................................................................................ 2-1

2.1 Introduction.................................................................................................................. 2-12.2 Discussion of Chemistry Regimes................................................................................ 2-22.3 Materials Integrity in the Reactor Coolant System ....................................................... 2-6

2.3.1 Corrosion Modes of Structural Materials ............................................................... 2-62.3.2 Effects of Chemistry Parameters........................................................................... 2-8

2.3.2.1 Dissolved Oxygen .......................................................................................... 2-82.3.2.2 Dissolved Hydrogen ....................................................................................... 2-92.3.2.3 pH ................................................................................................................ 2-102.3.2.4 Lithium.......................................................................................................... 2-132.3.2.5 Effect of Lithium-pH History on PWSCC Susceptibility.................................. 2-132.3.2.6 Chloride........................................................................................................ 2-152.3.2.7 Fluoride ........................................................................................................ 2-152.3.2.8 Sulfate.......................................................................................................... 2-162.3.2.9 Organics....................................................................................................... 2-172.3.2.10 Zinc ............................................................................................................ 2-17

2.4 Fuel Integrity Considerations ..................................................................................... 2-182.4.1 Chemistry Effects on Fuel Reliability ................................................................... 2-18

2.4.1.1 Cladding Corrosion....................................................................................... 2-182.4.1.2 Fuel Crud Deposition.................................................................................... 2-19

xviii

2.4.2 Effects of Chemistry Parameters....................................................................... 2-202.4.2.1 Dissolved Oxygen ........................................................................................ 2-202.4.2.2 Fluoride ........................................................................................................ 2-202.4.2.3 Aluminum, Calcium, Magnesium, and Silica................................................. 2-212.4.2.4 Suspended Solids ........................................................................................ 2-232.4.2.5 pH ................................................................................................................. 2-242.4.2.6 Lithium.......................................................................................................... 2-242.4.2.7 Zinc .............................................................................................................. 2-27

2.4.3 Alternate Cladding Material to Zircaloy-4 ........................................................... 2-292.4.4 Effect of pH on Fuel Axial Offset ....................................................................... 2-29

2.5 Radiation Field Control ............................................................................................. 2-312.5.1 Sources of Radiation Fields ................................................................................ 2-312.5.2 Effects of Chemistry Parameters......................................................................... 2-33

2.5.2.1 Corrosion Product Release .......................................................................... 2-342.5.2.2 Particulate Transport .................................................................................... 2-342.5.2.3 Soluble Transport......................................................................................... 2-352.5.2.4 Verification Tests and Plant Experience....................................................... 2-372.5.2.5 Plant Data .................................................................................................... 2-38

Uncoordinated versus Coordinated Chemistry Operation ..................................... 2-38Modified versus Coordinated Chemistry Operation ............................................... 2-39Elevated Lithium versus Coordinated Chemistry Operation .................................. 2-40Comparison of Radiation Field Trends as a Function of Coolant Chemistry ......... 2-40

2.5.2.6 Zinc .............................................................................................................. 2-412.5.3 Summary........................................................................................................... 2-42

3 POWER OPERATION CHEMISTRY CONTROL RECOMENDATIONS.............................. 3-13.1 Introduction................................................................................................................. 3-13.2 Reactor Coolant System (RCS) pH Optimization......................................................... 3-1

3.2.1 Operating Recommendations................................................................................ 3-13.3 Control and Diagnostic Parameters ........................................................................... 3-113.4 Guideline Recommendations..................................................................................... 3-12

3.4.1 Goals................................................................................................................... 3-123.4.2 Corrective Actions ............................................................................................... 3-12

3.5 Definitions of Terms Used in the Guidelines .............................................................. 3-13

xix

3.5.1 Plant Status......................................................................................................... 3-133.5.2 Action Levels....................................................................................................... 3-13

3.5.2.1 Action Level 1............................................................................................... 3-133.5.2.2 Action Level 2............................................................................................... 3-143.5.2.3 Action Level 3............................................................................................... 3-14

3.6 Guideline Values for Chemistry Parameters During Power Operation....................... 3-153.6.1 Reactor Coolant System Guidelines ................................................................... 3-153.6.2 Special Considerations ....................................................................................... 3-19

3.6.2.1 Axial Offset Anomaly .................................................................................... 3-193.6.2.2 Extended Fuel Cycles .................................................................................. 3-203.6.2.3 Operation With Greater Than 2.2 ppm Lithium............................................. 3-203.6.2.4 Chemistry Program Modification .................................................................. 3-21

3.6.3 Primary System Makeup Water........................................................................... 3-223.7 Control and Diagnostic Parameters, Frequencies, and Limits for StartupChemistry.......................................................................................................................... 3-23

4 METHODOLOGY FOR PLANT-SPECIFIC OPTIMIZATION ............................................... 4-14.1 Introduction.................................................................................................................. 4-14.2 Primary Water Chemistry Variables and Options......................................................... 4-2

4.2.1 Program Objectives............................................................................................... 4-24.2.2 Parameters Impacting Structural Integrity ............................................................. 4-3

4.2.2.1 Chloride/Fluoride/Sulfate................................................................................ 4-34.2.2.2 Dissolved Oxygen Control.............................................................................. 4-44.2.2.3 Hydrogen........................................................................................................ 4-44.2.2.4 pHT (Lithium Control Parameter)..................................................................... 4-54.2.2.5 Zinc ................................................................................................................ 4-5

4.2.3 Parameters With Negligible Impact on Structural Integrity .................................... 4-54.2.3.1 Silica............................................................................................................... 4-64.2.3.2 Suspended Solids .......................................................................................... 4-6

4.2.4 Chemistry Control During Shutdowns and Startups .............................................. 4-64.2.4.1 Shutdown Chemistry Program........................................................................ 4-64.2.4.2 Startup Chemistry Program............................................................................ 4-7

4.3 Optimization Methodology ........................................................................................... 4-74.3.1 Optimization Process ............................................................................................ 4-7

xx

4.3.2 NEI SG Initiative Checklist .................................................................................... 4-7

5 REFERENCES .................................................................................................................... 5-1

A CALCULATION OF PHT AND DATA EVALUATION METHODOLOGIES..........................A-1

B CHEMISTRY CONTROL OF SUPPORTING SYSTEMS ....................................................B-1

C STATUS OF ENRICHED BORIC ACID (EBA) APPLICATION...........................................C-1

D CVCS FILTER AND RESIN OPTIMIZATION PRACTICES ................................................D-1

E OXYGEN AND HYDROGEN BEHAVIOR IN PWR PRIMARY CIRCUITS ..........................E-1

F CONSIDERATIONS FOR REFERENCING HIGH TEMPERATURE PHCALCULATIONS TO 300°C................................................................................................... F-1

xxi

LIST OF FIGURES

Figure 2-1 Schematic of the Problems with Optimizing RCS pH Control ................................ 2-2Figure 2-2 Schematic of Coordinated Chemistry Regime (at 300°C)...................................... 2-4Figure 2-3 Schematic of Elevated Chemistry Regime (at 300°C) ........................................... 2-4Figure 2-4 Schematic of Modified Chemistry (at 300°C) ........................................................ 2-5Figure 2-5 Schematic of Coordinated Chemistry at pHT 7.1 Consistent with Lithium

Below 3.5 ppm (at 300°C) ............................................................................................... 2-5Figure 2-6 The Effects of Oxygen and Chloride on the Stress Corrosion Cracking of

Austenitic Stainless Steel in High Temperature Water (9)............................................... 2-7Figure 2-7 Variation of Crack Growth Rate in Two Heats of Alloy 600 at 325°C vs. pH........ 2-12Figure 2-8 Variation of Crack Growth Rate in Two Heats of Alloy 600 at 325°C vs.

Lithium Concentration ................................................................................................... 2-12Figure 2-9 Increase in PWSCC Susceptibility as a Function of Li & B (Relative to Base

Point)............................................................................................................................. 2-16Figure 2-10 Zircaloy Corrosion versus Core Burn-Up - Industry Experience ........................ 2-19Figure 2-11 Solubility of Magnesium Silicates at 350°C........................................................ 2-21Figure 2-12 Measured Oxide vs. Rod Average Burnup at Millstone and North Anna ........... 2-25Figure 2-13 Radiation Field Trends in PWRs........................................................................ 2-32Figure 2-14 Variations in Iron Solubility from Core Inlet to Outlet as a Function of pHT at

the Core Average Coolant Temperature (Boron = 600 ppm, H2 = 35 cc/kg H2O) .......... 2-36Figure 2-15 Variations in Nickel Solubility from Core Inlet to Outlet as a Function of pHT

at the Core Average Coolant Temperature (B = 600 ppm, H2 = 35 cc/kg H2O) ............. 2-36Figure 2-16 Variation with pH of Average Activities of Co-60 and Co-58 on Steam

Generator Tubing at the End of MIT in-Pile Loop Tests ................................................ 2-38Figure 2-17 Steam Generator Channel Head Dose Rates (Comparison of Trends of

Plants in Group 3 - Plant Startup after 1981) Data for Four Plants are Shown asThree-Year Moving Mean.............................................................................................. 2-41

Figure 3-1 Schematic Representation of the PWR Primary Chemistry OptimizationProblem........................................................................................................................... 3-3

Figure 3-2a Schematic of Fuel Integrity Risk for General RCS pH Regimes* ......................... 3-6Figure 3-3 Variation of pH with Boron Concentration when Lithium is Maintained within

+10% (2000-1650 ppm Boron), +0.30 ppm (1650-719 ppm Boron) and +24% (719-0 ppm Boron) Control Band for pH300°C = 6.9 .................................................................. 3-11

xxiii

LIST OF TABLES

Table 2-1 Effect of Primary Water Chemistry on Alloy 600 Crack Growth Rates at325°C (37)..................................................................................................................... 2-11

Table 2-2 Effect of Lithium-pH History on PWSCC Susceptibility (Tave = 308°C)................... 2-14Table 3-1 Generic Principles for Optimization of Primary System pH(1) ................................... 3-4Table 3-2 Operational Status Modes .................................................................................... 3-13Table 3-3 Reactor Coolant System Power Operation Control Parameters(9) (Reactor

Critical) .......................................................................................................................... 3-16Table 3-4 Reactor Coolant System Power Operation Diagnostic Parameters (Reactor

Critical) .......................................................................................................................... 3-17Table 3-5 Source Water for Reactor Makeup(1) Diagnostic Parameters -- All Modes............. 3-22Table 3-6 Reactor Coolant System Cold Shutdown Control Parameters (Reactor

<250°F) ......................................................................................................................... 3-23Table 3-7 Reactor Coolant System Startup Control Parameters(1) (Reactor Subcritical

and >250°F) .................................................................................................................. 3-24Table 3-8 Reactor Coolant System Startup Chemistry Diagnostic Parameters (Following

Fill-and-Vent to Reactor Critical) ................................................................................... 3-25Table 4-1 Chemistry Control Program Approaches ................................................................ 4-9

1-1

1 INTRODUCTION AND MANAGEMENT

RESPONSIBILITIES

1.1 Introduction

The purpose of this document is to provide recommendations on primary chemistrycontrol to continue to maximize the long-term availability of PWR plants. The nuclearpower industry has compiled an excellent performance history in maintaining PWRprimary system component integrity. In addition, primary water chemistry controlcan and has been effectively used to control radiation field buildup on ex-core surfaces.

It is important that all levels of utility management understand that operation outsidethe recommended chemistry limits can increase the probability of reduced unitavailability. Such losses may be limited by controlling the magnitude and frequency ofabnormal reactor coolant system (RCS) chemistry conditions. A principal goal has beento minimize corrosion of RCS components. Effectively implementing these Guidelineswill support this goal, thereby maximizing the cumulative availability of the unit. Inaddition, the US nuclear power industry established a framework for increasing thereliability of steam generators by adopting NEI 97-06, Steam Generator ProgramGuidelines. This initiative adopts EPRI’s Water Chemistry Guidelines, including thisdocument, as the basis for a standardized chemistry program. Specifically, theinitiative requires that US utilities meet the intent of the EPRI Primary Water ChemistryGuidelines as of the first refueling outage after January 1999.

The focus of the NEI initiative is steam generator integrity. It is recognized that steamgenerators comprise a major portion of the primary system pressure boundary.However, many aspects of primary chemistry are designed to insure primary systemand fuel cladding ,integrity and minimize radiation fields. Therefore, these Guidelinesinclude parameters and monitoring requirements which are recommended forinclusion in the water chemistry program, but should not be considered as required tomeet the intent of the NEI initiative. Volume 1 addresses the specific requirements ofthe primary water chemistry program relative to the NEI initiative. Section 4 of thisvolume provides utilities with specific direction on what details of these Guidelinesmust be considered to meet the intent of the NEI initiative. In addition, a checklistproviding for assessing conformance to the requirements of the NEI initiative is also

Introduction and Management Responsibilities

1-2

provided in Section 4. Volume 2 of these guidelines address aspects of startup andshutdown chemistry practices which are not believed to impact SG tube integrity.Therefore, utilities need not meet the intent of Volume 2 to be in compliance with theNEI Initiative.

A specific “NEI requirement” of Volume 1 is the completion of a Strategic WaterChemistry Plan. This plan is intended to be a living document which is approved byutility management. It is recognized that development of the first version of this plan isa significant challenge for utilities. As such, the committee agreed that completion ofthe plan should be accomplished within a year of the issue date of these Guidelines.

This revision to the Guidelines incorporates operational concerns and issues that havedeveloped since publication of the PWR Primary Water Chemistry Guidelines - Revision 3(TR-105714) in November 1995. The new issues of most significance addressed in therevision are:

• Optimization of high-temperature coolant pH (pHT) regimes to meet the demands oflonger fuel cycles (i.e., higher lithium concentrations to coordinate with higherboron) has been addressed in further depth. The treatment of this topic considerseffects of the higher lithium on structural materials, fuel corrosion, axial offsetanomaly, and shutdown radiation fields. This issue is addressed in Sections 2and 3.

• A more quantitative evaluation has been provided of the effects of elevated pH andlithium on PWSCC of Alloy 600 components in Section 2.

• Limits on silica have been re-evaluated and are covered in Sections 2 and 3.

• An expanded discussion of options and experience with chemistry control duringshutdowns and startups has been provided in Volume 2.

• A new Section 4 has been added to these Guidelines to provide a methodology fordeveloping a plant-specific optimized primary water chemistry program. Inaddition, this section addresses the specific requirements of the NEI SG Initiative(97-06) with respect to primary water chemistry. One of the requirements of thissection made necessary by the NEI initiative is the formal documentation of therationale for deviations from these industry Guidelines.

1.2 Generic Management Considerations

One major element of these Guidelines is the need for every level of management tounderstand the importance of the action levels presented in the Guidelines sections andtheir potential impact on, and benefits to, the utility company. Since organizationalstructures vary, approaches to Guidelines implementation are expected to vary.


1-3

However, in all cases, utility-specific implementation policies and procedures shouldbe developed to assign the responsibilities for Guidelines implementation to specificpositions within the organization. These written policies and procedures generally:

a. State the need for the policy.

b. State the corporate goal regarding water chemistry and station operation.

c. Highlight corporate management support for the policy.

d. Assign responsibility for:

• Preparation and approval of procedures to implement the policy

• Assessment of program effectiveness in accomplishing the established goals

• Monitoring, analysis and data evaluation for the chemistry program

• Surveillance and review functions

• Corrective actions

e. Establish the authority to:

• Carry out procedures

• Implement corrective actions

• Resolve disagreements

Procedures and policies normally should contain the level of detail necessary forpersonnel at all levels to understand and carry out their responsibilities.

1.3 Communication and Training

It is important that management expects and supports effective communicationbetween personnel involved in the control of chemistry during startup, operation andshutdown. Undesirable chemical conditions can occur as a result of miscommunicationbetween personnel in Operations and Chemistry, especially during changing plantconditions. This can result in significant unanticipated increases in plant radiation doserates and unplanned radiation exposures.

Personnel who implement and support the Primary Water Chemistry Guidelines shouldreceive initial and periodic training commensurate with their program responsibilities.The training programs should include the following elements:


1-4

• A clear statement of the corporate policy regarding water chemistry control,including clarification of the impact of this policy on the variousresponsibility areas.

• Discussion of the adverse impact of poor chemistry control on materials andfuel integrity and radiation fields, as well as potential impacts on companyeconomic performance.

• Techniques for recognizing unusual conditions and negative trends,particularly for the station chemists and laboratory technicians. Potentialcorrective actions and their consequences should be thoroughly discussed.

• The interaction between system operations and chemistry.

1.4 Outage Planning and Coordination

Plans for reactor startups and shutdowns should incorporate the activities and timerequirements necessary to support the recommendations presented in Volume 2 ofthese Guidelines. Such an effort is expected to minimize station radiation fields. Theshutdown/startup chemistry program can also influence fuel performance andeconomics since it can affect deposits present on the fuel thereby impacting on claddingcorrosion and the tendency for developing AOA. Startup or shutdown “shortcuts”with respect to chemistry should be avoided without first assessing the possibleimpacts. To obtain optimum results, many plant organizations, including Chemistry,Radiation Protection, Maintenance, Operations, and Outage Planning must coordinatetheir efforts prior to and during the shutdown evolution.

2-1

2 TECHNICAL BASIS FOR THE NEED TO CONTROL THE

COOLANT CHEMISTRY IN PWRS

2.1 Introduction

The purposes of PWR primary coolant chemistry guidelines are:

• To assure primary system pressure boundary integrity,

• To assure fuel-cladding integrity and achievement of design fuel performance, and

• To minimize out-of-core radiation fields.

In certain instances, situations are encountered where chemistry conditions that areoptimum for achieving one goal can lead to a decreased level of achievement relative toother goals. Unfortunately, this is the situation encountered in defining an optimumPWR primary coolant chemistry program. Specifically, operation at elevated pH canreduce out-of-core radiation fields but the accompanying elevated lithiumconcentrations can lead to an increased likelihood of Alloy 600 cracking. (Recent data,described in Section 2.3, indicate that this effect is smaller than previously estimated.)In contrast, at a lower operating lithium concentration, the likelihood of Alloy 600integrity problems will be minimized but an increased buildup rate of radiation fieldsmay result. The fuel cladding situation is even more complicated. Specifically, heaviercrud deposits are expected on the cladding at low pH, possibly leading to increasedcladding failure rates. Higher pH operation minimizes the rate of crud deposition,while the possibility of accelerated cladding corrosion in regions of the core wheresubcooled boiling occurs increases as local lithium concentrations increase.

As a result of considerations such as the above, the Guidelines must define waterchemistry parameters to achieve a balance among the three goals, recognizing thatlower priority is assigned to radiation field control compared with that for materialsand fuel integrity goals. Figure 2-1 illustrates the dilemma in defining an optimumRCS pH (lithium/boron concentration relation) to meet the conflicting requirements ofpressure boundary integrity, fuels integrity, and radiation control.

Technical Basis for the Need to Control the Coolant Chemistry in PWRs

2-2

Figure 2-1Schematic of the Problems with Optimizing RCS pH Control

The chemistry guidelines presented in Section 3 represent the results of the committee'sattempt to define a generic program for PWR primary systems. The specific valuesrepresent the "optimized chemistry" program for PWR primary water systems based oncurrent knowledge. The technical bases for these recommendations are summarized inthis section.

2.2 Discussion of Chemistry Regimes

The following discussion of high temperature pH (pHT) regimes is being provided tosimplify the discussions of chemistry effects presented later in this section. Selection ofan appropriate operating regime is part of chemistry optimization, discussed in greaterdetail in Sections 3 and 4. Several pH control schemes are discussed below and shownschematically in Figures 2-2, 2-3, 2-4, and 2-5.

In the early 1980s, many plants began to employ Coordinated Chemistry (Figure 2-2),in which lithium is coordinated with boron to maintain a pHT of 6.9. More recently,some plants have increased the coolant steady-state pHT from 6.9 to 7.2-7.4 to minimizecrud deposition on the core and reduce radiation fields in out-of-core regions. Asdiscussed later, studies suggest that operating with pH300°C = 7.4 will minimizesolubility-related nickel ferrite precipitation in the core. In the late 1980s, several PWRsimplemented a 3.5 ppm Li (max.)/pHT = 7.4 (Elevated Lithium) scheme to reduce out-


2-3

of-core radiation fields (see Figure 2-3). The Elevated Lithium chemistry programswere terminated due to concerns over the potential effects of prolonged exposure to 3.5ppm lithium on primary water stress corrosion cracking (PWSCC) of Alloy 600 andZircaloy cladding corrosion. Instead, a Modified Chemistry (Figure 2-4) regime with~2.2 ppm lithium/pHT = 7.2-7.4 was adopted by many plants. Maintaining lithiumconcentrations at a maximum of 2.2 ppm during startup, however, can result in a pHT

<6.9 for plants operating with an initial boron concentration exceeding ~1200 ppm.Depending on the cycle length and fuel design, the startup boron concentration mayexceed 2000 ppm, which will significantly lower the pHT. Therefore in the ModifiedChemistry regime, the startup lithium and boron are coordinated to maintain pHT ~6.9until the required lithium concentration decreases to 2.2 ppm. At this point, the pHT ispermitted to increase with a constant 2.2 ppm lithium, until it reaches 7.2-7.4. It isimportant to note that extended cycle chemistry may require significant concentrationsof lithium for the early portion of the cycle, as shown in Figure 2-4.

Over the past few years, several PWR operators have experienced axial offset anomalies(AOA) associated with high peaking factor core designs. The AOA problem, if notcontrolled, may require power reductions and/or plant shutdown. The mechanism ofAOA is not completely understood. However, it appears that AOA is associated withthe local concentration of boron in crud deposits on fuel rods with significant subcooledboiling. Crud deposits enhance the local concentration factor on the fuel and can leadto increased fuel surface cladding temperatures. For this reason, reducing the sourceterm for crud (e.g., minimizing corrosion rates of system materials) is consideredespecially important for plants susceptible to AOA. Corrosion product release rates arereduced as the pHT is increased from 6.9 to 7.4. The change in release rate becomes lessdependent on pHT as the pHT approaches 7.4. This dependence is observed for bothstainless steel and Alloy 600. This effect is more pronounced for Alloy 600, which isimportant because of the large surface area of SG tubing. Additionally, the corrosionand possibly the release rates are expected to be highest at the beginning of the cycleand decrease as protective oxides on system materials stabilize. On this basis, corrosionproduct generation and transport are expected to be greatest during the earliest parts ofthe fuel cycle when the pH is lowest under current practices. To address this situation,several plants have started to use a Coordinated Chemistry at Elevated pH regime (seeFigure 2-5) in which the pHT is maintained as close to constant and as high as practicalthroughout the full cycle, while still giving due consideration to PWSCC and fuelreliability concerns.


2-4

pH = 6.9 + 0.3 ppm Li

pH = 6.9

B, ppm

Li, ppm

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.002400 2300 2200 2100 20001900 18001700 1600 1500 1400 13001200 1100 1000 900 800 700 600 500 400 300 200 100 0

Note: Changes in control band width at high and low lithiumper Table 3-1 are not shown.

T

T

Figure 2-2Schematic of Coordinated Chemistry Regime (at 300°C)

B, ppm

Li, ppm

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.002400 2300 22002100 2000190018001700 1600 150014001300 1200 1100 1000 900 800 700 600 500 400 300 200 100 0

pH = 7.4 - 0.15 ppm LipH = 7.4

pH = 7.4 + 0.15 ppm Li

Li = 3.35 ppm - 0.15 ppmLi = 3.35 ppm

Li = 3.35 ppm + 0.15 ppm

Figure 2-3Schematic of Elevated Chemistry Regime (at 300°C)


2-5

B, ppm

Li, ppm

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.002400 2300 2200 21002000190018001700 1600 150014001300 120011001000 900 800 700 600 500 400 300 200 100 0

pH = 6.9 + 0.3 ppm LipH = 6.9

constant 2.35 ppm Liconstant 2.20 ppm Li

constant 2.05 ppm Li

pH = 7.4 + 0.15 ppm Li pH = 7.4

pH = 7.4 - 0.15 ppm Li

T

T

T

T

T


Figure 2-4Schematic of Modified Chemistry (at 300°C)

B, ppm

Li, ppm

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.001800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 200 100 0

pH = 7.0

pH = 7.1 pH = 7.2 pH = 7.3 pH = 7.4

pH = 6.9

1.5 ppm Li

3.5 ppm Li

2.2 ppm Li


T

T

T

TTT

Figure 2-5Schematic of Coordinated Chemistry at pHT 7.1 Consistent with Lithium Below 3.5ppm (at 300°C)


2-6

2.3 Materials Integrity in the Reactor Coolant System

Austenitic stainless steels (e.g., Type 304, 316, and A-286) and nickel-base alloys (e.g.,Alloys 600, 690, X-750, and 718) are major construction materials used in the reactorcoolant system (RCS). Typical applications of Alloy 600 in the RCS include steamgenerator tubes and plugs, pressurizer instrument nozzles, pressurizer heaters andheater sleeves, piping safe ends, and hot leg instrument nozzles. In addition, Alloy 600penetrations are also used in the closure head and bottom head of the reactor pressurevessel. High strength alloys X-750, 718, and A-286 are extensively used in coreinternals as bolts, springs, and guide pins. Corrosion considerations related to thesematerials are discussed below.

2.3.1 Corrosion Modes of Structural Materials

Water chemistry impacts both general corrosion and stress corrosion cracking (SCC) ofRCS materials. General corrosion is an oxidation process which occurs relativelyuniformly over a material surface. If the environment is properly controlled, the oxidefilm formed will lead to a reduction in the rate of corrosion until a steady-statecondition is reached between oxide film thickness (protective) and the generalcorrosion rate. Some of the corrosion products released from system surfaces will formdeposits on the fuel elements. Activation products of species present in such deposits(e.g., Co-58, Co-60) and their release, transport, and deposition on out-of-core surfaceslead to increased plant radiation levels, which are of particular concern during outages.

Minimization of oxygen and halide concentrations in the RCS is necessary to preventSCC of austenitic stainless steels (see Figure 2-6). There is substantial literaturedescribing the susceptibility of other RCS materials such as alloys 600, X-750, 718 andA-286 to this type of corrosion (4-10). This literature indicates that susceptibility to SCCis not significantly affected by the variations in halide and sulfate concentrations likelyto be experienced in the RCS as a result of the low ECP of the system alloys in theoxygen free environment. However, small concentrations of oxygen can significantlyincrease the ECP and increase alloy susceptibility. The strong influence of oxygen isillustrated by the increased susceptibility of these alloys to SCC in BWR coolant ascompared to PWR coolant. This increased susceptibility is attributed mostly to theabout 200 ppb oxygen in the BWR coolant, although differences in pH and temperatureare also factors.

Numerous instances of localized corrosion, specifically SCC, have occurred. PWSCC ofmill-annealed Alloy 600 steam generator tubes and plugs has been well documentedand remedial actions have been developed to mitigate this problem (4,5). Current datasuggest that coolant chemistry is a second-order effect in the PWSCC process (11).Other factors that more directly contribute to PWSCC are high coolant temperature,high applied and residual stresses, cold work, and a susceptible alloy microstructure.


2-7

Figure 2-6The Effects of Oxygen and Chloride on the Stress Corrosion Cracking ofAustenitic Stainless Steel in High Temperature Water (9)

Mill-annealed Alloy 600 is available in a wide variety of metallurgical conditions.Annealing Alloy 600 at higher temperatures is expected to produce a microstructuremore resistant to PWSCC (1). Associated with higher temperature annealing are thefollowing properties: 1) a lower yield strength which determines the upper limit forresidual stresses, 2) larger grain size, and 3) a higher population of grain-boundarycarbides which appear to impart resistance to PWSCC (2). In an effort to improve theSCC resistance of Alloy 600, a process known as thermal treatment has been utilized.In general, thermal treatment involves heating the material to 1300°F for 15 hours. Thisprocess tends to improve resistance to SCC by producing a semi-continuous grainboundary chromium carbide precipitate, significantly reducing residual stresses andslightly reducing yield strength. In spite of thermal treatment, the SCC resistance of thefinal product is strongly influenced by the final anneal prior to thermal treatment (3).


2-8

Many replacement steam generators have been tubed with Alloy 690 and Alloy 800,which are significantly more resistant to PWSCC than Alloy 600.

More recently, Alloy 600 penetrations in PWR pressurizers and reactor vessel headshave experienced cracks and leaks caused by PWSCC. Such cracking has occurred atboth U.S. and foreign plants (6, 7). Plants with stainless steel penetrations in thepressurizer have not experienced cracking. In several plants, PWSCC has occurredafter times ranging from one to twenty years of plant operation; whereas in other cases,no PWSCC has been detected after twenty or more years of operation. Factors such aslocal residual stresses produced by welding, surface cold work and materialmicrostructure have a significant effect on time to cracking. Some plants have replacedreactor vessel head penetration materials with Alloy 690 because of its superiorresistance to PWSCC. The NRC has issued Generic Letter 97-01 requesting that utilitiesdescribe their plans to address degradation of vessel head penetrations.

2.3.2 Effects of Chemistry Parameters

The key chemistry parameters requiring control to minimize general corrosion and SCCare discussed below. A recent EPRI study evaluated the literature data on the effects ofdissolved hydrogen, pH and lithium on PWSCC of Alloy 600 (11). The effect of boronwas also considered implicitly by its correlation with pH at fixed lithium concentration.The results of this evaluation are discussed in Sections 2.3.2.2 to 2.3.2.5.

2.3.2.1 Dissolved Oxygen

Minimization of coolant oxygen concentrations will lead to minimization of both SCCand general corrosion in the RCS (12). Dissolved oxygen concentrations can becontrolled during plant heat-up by venting or vacuum filling followed by the use ofhydrazine or hydrogen for residual oxygen scavenging.

Revision 2 of these Guidelines stated that for typical PWR flux spectra at 1100 ppm B,Burns showed that hydrogen concentrations of 14-15 cc (STP) H2/kg H2O scavengeradiolytically produced oxygen (13). An adaptation of the Jenks model (14) forcomputing the rate of formation of oxidizing species by radiolysis under PWRconditions suggests that the scavenging of the aggressive hydroperoxy (HO2⋅) radical isachieved if 15-20 cc (STP) H2/kg H2O are present in the coolant during power operation(15). Studies at a 1300 MW PWR indicated that the dissolved hydrogen concentrationof 7 cc/kg predicted by earlier studies to prevent formation of oxidizing species in thePWR is conservative (16). This conclusion is supported by calculations made byChristensen (17). Christensen concluded that at a hydrogen concentration of 5 cc/kgthe oxidizing species concentrations are only slightly higher than at 15 cc/kg and that,although the recommended range of 25-50 cc/kg may be necessary at lowtemperatures, a lower value would be sufficient at normal operating temperatures.


2-9

A recent analysis of oxygen and hydrogen behavior in PWR primary circuits wascompleted by Garbett et al., and is summarized in Appendix E. This analysis indicatesthat the levels of hydrogen required to suppress radiolysis during power operation arein the range of <1 - 5 cc/kg, i.e., well below the specified 25 cc/kg minimum, and thatthis remains true even when subcooled boiling is occurring. Hydrogen peroxide is thedominant oxidizing species formed in the core at all hydrogen concentrations, withoxygen and other oxidizing radicals being present at much lower concentrations.Hydrogen peroxide should dominate the solution and heterogeneous reactions both in-core and out-of-core and there is no reason to expect that the radicals are moreaggressive towards surface reactions. The analysis indicates that levels of oxidizingspecies downstream of the core are very low and are independent of the oxygen levelof the water coming into the core. This is also true during a shutdown as long as ahydrogen excess is maintained.

2.3.2.2 Dissolved Hydrogen

Hydrogen is added to maintain reducing conditions in the primary coolant to minimizeprimary system corrosion. Oxidizing conditions would lead to increased formationand transport of corrosion products, higher radiation fields, reactivity anomalies, crudbuildup on fuel, increased corrosion of fuel rods, and increased susceptibility of somestructural materials to SCC. As noted in section 2.3.2.1, the computations of productionrates of oxidizing species by radiolysis suggest a dissolved hydrogen concentration ofsignificantly less than 15 cc/kg is sufficient to scavenge the oxidizing species under alloperating conditions. Since oxygen can also be added to the coolant from othersources, an excess inventory of hydrogen must be maintained while the reactor is atpower. Revision 0 of the Guidelines set a range of 25-50 cc/kg to provide a marginagainst oxidizing conditions and to facilitate operational control.

Revision 1 presented evidence from laboratory studies that dissolved hydrogenincreased the susceptibility of some mill-annealed Alloy 600 steam generator tubing toSCC initiation in high-temperature (360°C) water (18, 19). In several tests, the time-to-cracking was inversely proportional to the dissolved hydrogen concentration. For thisreason, in Revision 1 a tighter range of 25-35 cc/kg was suggested for plants withsusceptible mill-annealed Alloy 600, and this recommendation was repeated inRevision 2.

In an EPRI project Eason (11) statistically analyzed published data from several PWSCCstudies on Alloy 600 (20-25) to evaluate the effect of dissolved hydrogen and othervariables on PWSCC initiation. In some cases, the investigators provided EPRI detailedtest data to facilitate the multivariable statistical analysis. The analysis considered theeffects of environmental variables as well as heat, heat treatment, specimen type andstress state on the observed times to cracking. The temperature range was 320-330°C ascompared to up to 360°C in earlier studies. The results of this analysis showed that


2-10

stress related and material related effects were dominant and water chemistry effectswere secondary. The analysis was designed to block out the dominant stress andmaterial effects as well as possible, but considerable uncertainty remains in the waterchemistry correlation that was produced. This comment also applies to the relateddiscussion of pH and lithium and the model in Sections 2.3.2.3 through 2.3.2.5 below.

There was an increase of approximately 10% in the Weibull characteristic life η (η =time to crack of 63.2% of the population) at dissolved hydrogen concentrations of about15 cc/kg relative to the characteristic life with hydrogen concentrations of 25-50 cc/kg.The study concluded that dissolved hydrogen in the range of 25-50 cc/kg did not havea significant effect on PWSCC initiation times, as measured by the Weibullcharacteristic life η. There are only limited data available on the effect of hydrogen oncrack growth rates (26). The data do not show a significant effect of hydrogen in the 25-50 cc/kg range on Alloy 600 crack growth rates. Therefore, the recommendation inRevision 3 for a tighter range of 25-35 cc/kg for plants with susceptible material is nolonger justified, and a range of 25-50 cc/kg can be applied to all plants. However, ifplants are experiencing significant PWSCC degradation of Alloy 600 components theymay wish to consider operating at a lower hydrogen concentration of 15-25 cc/kg tomitigate PWSCC. Operation in this range may require increased monitoring to ensurethat hydrogen does not fall below the lower limit. Plants considering operation in thisrange should conduct an evaluation to address the effect of the lower hydrogenconcentration on fuel cladding deposition and corrosion, while recognizing that datasupporting the benefit of this approach are limited. Any decision to use lowerhydrogen levels should also recognize that there are some uncertainties with regard tothe needed minimum level of hydrogen that is required, e.g., to protect againstenhanced radiolysis due to the effects of impurities (103).

2.3.2.3 pH

Evaluation of the effect of pH on PWSCC crack initiation is complicated by thecorrelation between pH and temperature which makes it difficult to compare studies atdifferent temperatures. The statistical analysis discussed in 2.3.2.2 indicates that in thetemperature range of 320-330°C, there is a moderate decrease (~ 14%) in thecharacteristic life η when the pHT is increased from 7.0 to 7.4 (11). There is little effectin the pHT range of 7.4-7.8. All of these pH values are given at the temperature of thetests, which was 320-330°C. At 300°C the corresponding pH values would beapproximately 0.45 units lower.

The effect of water chemistry on PWSCC crack growth was also investigated in an EPRIsponsored study (37). Tests were conducted on two heats of Alloy 600 in selectedboron-lithium concentrations corresponding to beginning, middle and end of cyclevalues for coordinated and modified boron and lithium control. The chemistryconditions tested and the results are shown in Table 2-1 and Figures 2-7 and 2-8. The


2-11

figures show that crack growth rate was not systematically dependent on waterchemistry variations (including pH and lithium) within the PWR water chemistryguidelines. The effect of chemistry on crack growth was second order compared to theheat to heat variability.

The overall conclusion that can be drawn from these studies is that pH in the operatingrange has a relatively small effect on PWSCC of Alloy 600. The possible effect of pH onshutdown radiation fields is discussed in Section 2.5.

Table 2-1Effect of Primary Water Chemistry on Alloy 600Crack Growth Rates at 325°C (37)

Heat 20Specimen Boron ppm Li ppm pHT* CGR, m/s

20-17 1814 2.9 7.23 9.6E-1220-1 1190 2.0 7.27 5.4E-1120-4 315 0.5 7.17 5.6E-1220-12 309 1.6 7.66 8.7E-1220-18 1200 2.0 7.27 9.4E-12

Heat 69Specimen Boron ppm Li ppm pHT* CGR, m/s

69-3 1194 2.2 7.31 1.5E-1069-9 1905 3.2 7.25 5.0E-1169-10 335 1.7 7.66 1.9E-1069-11 1200 2.0 7.27 2.2E-10

* pHT calculated using chemWORKSTM


2-12

Alloy 600 Crack Growth Rate at 325°C vs. pHT at 325°C

1.0E-12

1.0E-11

1.0E-10

1.0E-09

7.1 7.2 7.3 7.4 7.5 7.6 7.7

pH at 325°C

Cra

ck G

row

th R

ate,

m/s

Heat 20

Heat 69

Figure 2-7Variation of Crack Growth Rate in Two Heats of Alloy 600 at 325°C vs. pH

Alloy 600 Crack Growth Rate at 325°C vs. Li

1.0E-12

1.0E-11

1.0E-10

1.0E-09

0 1 2 3 4

Li, ppm

Cra

ck G

row

th R

ate,

m/s

Heat 20

Heat 69

Figure 2-8Variation of Crack Growth Rate in Two Heats of Alloy 600at 325°C vs. Lithium Concentration


2-13

2.3.2.4 Lithium

As noted earlier, PWSCC has occurred principally in highly-stressed regions (e.g., U-bends and tubesheet expansion transitions) in steam generators with susceptible Alloy600 tubing and in Alloy 600 tube plugs and in vessel head and pressurizer penetrations.(Refer to Section 2.3.1 for a qualitative description of some metallurgical conditionswhich influence the susceptibility of Alloy 600 to PWSCC.) Peening and/or stress reliefefforts have been applied as remedial actions to PWSCC in steam generators. Theinfluence of elevated lithium on the initiation of PWSCC in susceptible tubing has beenevaluated (11). A statistical evaluation of laboratory data reported in references 18-25showed an approximately 20% decrease in the characteristic life η with increasinglithium concentration from 0.7 to 3.5 ppm and little additional effect of lithium above3.5 ppm (11). Thus, the effect of lithium is small compared to more dominant effects ofstress, heat-to-heat variations and temperature, and only becomes a significant factor iflong-term operation at elevated lithium concentrations (e.g., 3.5 ppm lithium) isconsidered or if relatively small increases in susceptibility to PWSCC (i.e., decreases incharacteristic life η) are of concern at the plant.

The effect of water chemistry on crack growth has been summarized in the previoussection. Tests were conducted with boron and lithium concentrations (and thus pH)typical of coordinated and modified chemistry control. The data show (Figures 2-7 and2-8) that crack growth rates in both the heats tested were not systematically affected bychanges in lithium concentrations and pH within the PWR chemistry guidelines.Therefore, these data support the conclusion that chemistry regimes with initial lithiumconcentrations up to 3.5 ppm should not cause a significant increase in Alloy 600 crackgrowth rates.

2.3.2.5 Effect of Lithium-pH History on PWSCC Susceptibility

In order to evaluate the possible influence of different lithium - pH histories onsusceptibility of Alloy 600 to PWSCC, a susceptibility model based on a statisticalevaluation of Alloy 600 PWSCC data was used (11) to predict characteristic life as afunction of lithium, pH, hydrogen concentration and material/stress state. The modelconsists of the following equation:

η= -349Li - 40.1H2 - 1980pH + kLAB/RUB where (eq. 2-1)

η = time in hours to 63.2% failed of items being modeledLi = lithium concentration, ppm, for Li < 3.5 ppmH2 = hydrogen concentration, cc/kg (STP), for H2 < 25 cc/kgpH = pH at 320-330°C, for pH < 7.4kLAB/RUB = constant reflecting material susceptibility and stress state


2-14

The limits on Li, H2, and pH listed above, (3.5 ppm, 25 cc/kg., and 7.4 at T= 320-330°C)must be strictly observed, as the equation is invalid above those values. In thisevaluation, the intent was to examine the effects of lithium and pH. For this reason, thehydrogen concentration was held constant at a value of 25 cc/kg, and thematerial/stress constant was set at 21,800 hours, a mean value for the test specimensevaluated in the statistical study (11). The plant Tave was assumed to be 308°C and itshot leg temperature was assumed to be 325°C.

A typical 18 month fuel cycle length and boron concentration profile were used for theevaluation, with a cycle length of 469 days and a maximum boron concentration of 1400ppm. The characteristic life η was determined for four different lithium - pH histories:

(1) Coordinated pH 6.9, with constant pH 6.9 for the full cycle (Reference Case,Case 1)

(2) Coordinated pH 7.1, with constant pH 7.1 for the full cycle

(3) Modified pH 6.9-7.4, with constant pH 6.9 to lithium 2.2 ppm, constant lithiumat 2.2 ppm to pH 7.4, constant pH 7.4 to end of cycle

(4) Modified pH 7.1-7.2, with constant pH 7.1 to lithium 2.2 ppm, constant lithiumat 2.2 ppm to pH 7.2, constant pH 7.2 to end of cycle

For each chemistry history, the cycle was broken into about 13 time segments. For eachtime segment the value of characteristic life η was calculated, and then a time weightedaverage value of η was calculated for the full cycle. The average value of η for the fourchemistry histories is shown in the table below. The table also shows the percentdecrease in characteristic life η compared to the Reference Case, Case 1.

Table 2-2Effect of Lithium-pH History on PWSCC Susceptibility (Tave = 308°C)

Case Chemistry History ppm-Lidays

CharacteristicLife η hours

Percent Decrease in ηvs. Case 1

1 Coordinated 6.9 555 6186 - - -2 Coordinated 7.1 900 5539 12%3 Modified 6.9-7.4 1002 5552 11%4 Modified 7.1-7.2 1101 5454 13%

As shown in the above table, variations in pH - lithium water chemistry historieswithin the normal range used in the industry are predicted to have a small adverseeffect on susceptibility to PWSCC, in the range of 10 to 15% percent. To verify that theabove conclusion remains true for longer cycles, the same procedure was used to


2-15

evaluate the characteristic life for a nominal 20 month cycle. The cycle length was 588effective full power days, the maximum boron concentration was 1563 ppm, and thepH program was a modified pH 6.9 - 7.4 program. The calculated characteristic life ηwas 5540 hours, essentially the same as the 5552 hours for the 18 month modified pH6.9-7.4 program.

The above methodology has been used to evaluate the predicted median effect onPWSCC susceptibility over the Li - B plot, as shown in Figure 2-9. The iso-susceptibility lines on the plot indicate the increase in susceptibility to PWSCC relativeto the susceptibility for the following base point: lithium - 1.38 ppm, boron - 800 ppm,pH308°C - 6.9. As a further aid to assessing effects of chemistry on susceptibility toPWSCC, the methodology described above for calculating a cycle-averagedcharacteristic life η is being added to the pH Calculator in chemWORKS.

As discussed in the preceding paragraphs, statistical evaluation of available test dataregarding effects of chemistry on PWSCC indicates that the median effect of chemistryon the rate of PWSCC is modest, with expected effects of about 15% or less over a fuelcycle for changes in chemistry within the normal range used in the industry. However,the same statistical evaluation indicates that there is significant variability in responsesto chemistry (11). This variability needs to be considered when significant changes inchemistry are being evaluated.

2.3.2.6 Chloride

Chloride induced SCC occurs when austenitic stainless steels are exposed to chlorideions in the presence of oxygen in high-temperature water. A review of chlorideinduced SCC is given in reference (9). The inter-relationship between chloride andoxygen concentrations on SCC of austenitic stainless steels is shown in Figure 2-6. Thefigure suggests that the concern for SCC is minimal at typical RCS chemistryconditions.

2.3.2.7 Fluoride

Literature information about fluoride-induced SCC of austenitic stainless steelsapplicable to PWRs is limited. Whyte and Picone indicated that only sensitizedstainless steel is susceptible to fluoride-induced SCC (27). The lowest concentration offluoride ions tested that led to SCC was 10 ppm. When the test solutions containedboric acid, no SCC was detected. A much lower value for fluoride is in the TechnicalSpecifications for many stations.


2-16

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0200400600800100012001400160018002000Boron Concentration (ppm)

Lith

ium

Con

cent

ratio

n (p

pm)

pH6.9

pH7.0

pH7.4

pH7.3

pH7.2

pH7.1

1.0

1.05

1.1

1.15

1.2

1.0 1.05

1.1

1.15

1.2

1.25

Base PointLi = 1.07 ppmB = 800 ppmpH308°C = 6.9

Notes: (1) The relative susceptibility lines show the ratio of time to crack at the base point divided by that at the indicated point.

(2) Dashed lines show extrapolation beyond limits of model assuming (a) susceptibility for pH325°C > 7.4 equals that at pH325°C = 7.4, and (b) susceptibility for Li > 3.5 ppm equals that of Li = 3.5 ppm.

pH lines are for 308oC.

Figure 2-9Increase in PWSCC Susceptibility as a Function of Li & B (Relative to Base Point)

2.3.2.8 Sulfate

Laboratory studies have confirmed that sulfate is at least as detrimental as chloridewith respect to SCC of stainless steels in the high purity environments of BWRs (28).The relationship of sulfate and SCC in the PWR environment has been less wellstudied. At two PWRs, primary-side cracking of Alloy 600 was attributed to highconcentrations of reduced sulfur-bearing species (species containing sulfur in anoxidation state <6) (29, 30). This attack is believed to have occurred during cold


2-17

shutdown conditions at one plant and during power operation at the other (31, 32).During power operation, low level resin ingress can result in sulfate (or possibly othersulfur species) contamination of the RCS and possibly contribute to IGA. For plantswith filters downstream of CVCS ion exchangers, the likelihood of such an event isreduced. Due to the severity of sulfur compound-induced IGA and the lack ofunderstanding of the role of sulfate or reduced sulfur species in the corrosion process,Revision 3 of the Guidelines adopted sulfate as a control parameter.

In 1994 circumferential intergranular attack (IGA) was discovered in a control rod drivemechanism (CRDM) penetration at the Jose Cabrera plant in Spain. The IGA wasfound to have occurred in heavily sensitized Alloy 600 material. The probable cause ofthe IGA was determined to be the ingress of large amounts of resin in 1980 and 1981 (40and 200-300 liters, respectively) leading to high sulfate levels and low pH, coupled withthe presence of heavy sensitization of the Alloy 600 material (33). (NSSS estimates arethat about 30 liters of resin would result in a sulfate concentration of about 15 ppm (34,35).) NRC Generic Letter 97-01 requested utilities to review their water chemistryhistories to determine if resin ingress or other sulfate incursions had occurred similar tothe events at the Jose Cabrera plant. These reviews were completed, and responseswere filed with the NRC’s Public Document Room (34, 35, 36). In summary, it wasdetermined that no domestic plants have experienced resin or sulfate intrusions withmagnitudes similar to those at Jose Cabrera, and that monitoring of RCS waterchemistry in accordance with the Guidelines, i.e., by periodic sulfate analyses andLi/B/pH/conductivity data consistency checks, provides satisfactory protectionagainst large undetected ingress of resins and sulfates.

2.3.2.9 Organics

Organic materials and their decomposition products can enter the RCS from makeupwater, resin beds, or from boric acid impurities. There are no known detrimentaleffects of these organic species on fuels and materials integrity or plant radiation fields,except for the sulfur species attack discussed above. Organic materials are known tohave a short life in the temperature and neutron/gamma flux of the RCS. TheseGuidelines include suggested operating limits for sulfur or halogens that can beconstituents of organic species, and operators are encouraged to prevent the ingress oftrace levels of such species. There are no limits presented for organics.

2.3.2.10 Zinc

Previous work suggests that low concentrations of zinc in the PWR primary coolantcould reduce PWSCC of Alloy 600 (38, 39). A cooperative program was undertaken byEPRI and the Westinghouse Owners Group to evaluate the effectiveness of zinc inmitigating Alloy 600 PWSCC and radiation fields and to assess the effect of zinc oncorrosion of fuel cladding. The program included evaluation of zinc addition at Farley-


2-18

2. The scope of the program at Farley-2 included pre- and post- zinc measurements ofPWSCC of steam generator tubes during Cycle 10 (38, 39). At Farley-2, zinc was addedas natural zinc acetate after 190 days into Cycle 10. The coolant zinc concentration wasmaintained in the range of 35-45 ppb for approximately nine months. At the end ofCycle 10, evaluations were made of PWSCC of steam generator tubes. However,because of inspection uncertainty and the limited duration of zinc addition, furtheroperation with zinc is required before its effect on PWSCC can be evaluated withconfidence.

No zinc was added during Cycle 11 at Farley-2. Farley-2 resumed zinc addition inCycle 12. However the period of zinc addition during Cycle 12 was too short (3months) to assess its effect on PWSCC.

Diablo Canyon Unit 1 commenced adding natural zinc in June 1998, and plans to add itfor approximately eight months. The effect of zinc on PWSCC (as well on as radiationdose rates and fuel cladding corrosion) will be monitored under an EPRI-PG&Esponsored program.

2.4 Fuel Integrity Considerations

The zirconium alloy fuel rod cladding forms a barrier against the release of theradioactive fission products formed during operation. Maintaining cladding integrity,therefore, is a major objective of the plant operator and licensing authorities.

2.4.1 Chemistry Effects on Fuel Reliability

2.4.1.1 Cladding Corrosion

At normal operating temperatures, the zirconium alloy corrosion reaction can bedescribed by:

Z r + 2H O Z r O + 4H2 2→ (eq. 2-2)

The major effect of corrosion on cladding integrity results from two related processes:1) the formation of a ZrO2 layer causes cladding thinning and 2) the absorption ofhydrogen (the atomic hydrogen produced by corrosion) by the zirconium alloyembrittles the cladding. Both phenomena degrade cladding integrity. Factors affectingthe above reaction include temperature (zirconium alloy oxidation as shown in eq. 2-2is most significant at elevated temperatures; embrittlement is of most concern at lowertemperatures, e.g., during fuel handling), the microstructure of the cladding material,local boiling, and coolant chemistry.


2-19

The rate of zirconium alloy corrosion is controlled by the temperature of themetal/metal-oxide (Zr/ZrO2) interface. Since the cladding transfers heat to the coolant,the interface temperature during operation is always higher than that of the coolant.As the oxide grows thicker, the interface temperature increases, which in turn increasesthe corrosion rate. This effect is the major factor leading to accelerated corrosion withburnup, as shown for Zircaloy-4 in Figure 2-10. With thick oxide films and high heatflux, the corrosion rate increases further and the corrosion can cause premature failure.Besides temperature, cladding material plays a major role in the corrosion process, andthere is also in-reactor enhancement of the corrosion rate relative to autoclave data.Both out-of-pile and in-pile tests confirm that the corrosion rate can be reducedsignificantly through careful control of fuel cladding chemical composition and themanufacturing process (40). Because the bulk of industrial experience is with Zircaloy-4, the remainder of this section is written with Zircaloy-4 as the assumed fuel claddingmaterial. It is acknowledged, however, that other alloys are being used as claddingand for fuel structural components, and it is believed that the coolant chemistry issuesthat apply to Zircaloy-4 also apply, at least qualitatively, to these newer alloys. Thesealternate alloys are further discussed in Section 2.4.3.

100

80

60

40

20

10 20 30 40 50Local Burn-Up (GWd/MtU)

Oxi

de T

hick

ness

(µm

)

Figure 2-10Zircaloy Corrosion versus Core Burn-Up - Industry Experience

2.4.1.2 Fuel Crud Deposition

Corrosion products deposited on fuel surfaces will lead to increased claddingtemperatures and corrosion rates. Such deposits, as characterized during fuelexaminations, have been shown to contain non-stoichiometric nickel ferrites,

NixFe3-xO4, where typically 0.4<X<0.8 (41, 42). Higher nickel/iron ratios also have beenobserved, particularly on fuel that experienced subcooled boiling (43, 44). Values of X


2-20

of 1.5 or greater have been reported (43, 44). When X is greater than 1 (Ni/Fe ratio>0.5) excess Ni is present over that contained in the nickel ferrite lattice. This excessnickel exists either as NiO or elemental Ni. Lower-than-normal values of pH andhydrogen have been observed to increase the surface concentration of deposits on thecladding, as discussed later. Subcooled boiling has been found to enhance cruddeposition in some plants, and heavy deposits are believed to have initiated subcooledboiling at other plants. Because crud deposits can accelerate corrosion to the extentthat fuel rods fail, one of the primary goals of coolant chemistry is to minimizedeposition on the fuel.


The key chemistry parameters requiring control to minimize fuel integrity concerns arediscussed in the following sections.

2.4.2.1 Dissolved Oxygen

Oxygen ingress into the RCS occurs via the make up water. At very low hydrogenconcentrations (< 5 cc/kg), oxygen concentrations will increase due to radiolysis.Within the range of oxygen levels encountered in PWRs, the effect of oxygen oncorrosion of Zircaloy is minimal, and the greater concern is formation of systemcorrosion products and transport of these species to the core where they deposit on fuelsurfaces. As discussed in Volume 2, dissolved oxygen concentrations can be controlledduring plant heat-up by hydrazine, hydrogen and/or venting. During poweroperation, the concentrations of oxygen and oxidizing species are controlled by theaddition of hydrogen to the reactor coolant. Specific examples exist (43, 46) where lossof hydrogen control led to enhanced crud deposition. Since hydrogen rapidlycombines with oxygen residuals in the core, discernible differences between plantswhich operate with deaerated makeup water and those that operate with much higherconcentrations (e.g., 1-2 ppm) of oxygen in the make up water have not been observed.Nevertheless, it is considered desirable to deaerate makeup water if equipment ispresently installed. However, capital expenditure on new deaeration equipment isdifficult to justify.

2.4.2.2 Fluoride

A coolant fluoride specification of <2 ppm should provide protection against fluorideattack in PWR systems (47). A much lower value for fluoride is contained in theTechnical Specifications of most PWRs.


2-21

2.4.2.3 Aluminum, Calcium, Magnesium, and Silica

These impurities may be observed intermittently in the reactor coolant as a result ofmakeup water, refueling water, or boric acid contaminants. None of these elementshave a direct effect on Zircaloy corrosion. However, many oxides and silicates ofaluminum, calcium, and magnesium have negative temperature coefficients ofsolubility, i.e., they will deposit preferentially on the fuel rods (48), and could increasecorrosion of the Zircaloy. In particular, the zeolite minerals may cause densification ofporous crud deposits, especially in plants where boiling occurs. There are no data thatsuggest silica at concentrations <1000 ppb in the absence of the cationic species willform deposits. (Silica solubility is >200 ppm at RCS conditions.) However, calculationsperformed via the MULTEQ Code, as shown in Figure 2-11, suggest that highconcentrations of silica (e.g., 1-10 ppm) will precipitate low concentrations of zeolite-forming cations. If aluminum, calcium or magnesium are present, precipitation willoccur rapidly, and high concentrations of the cations will not likely be detected in thecoolant. Therefore, periodic monitoring of the concentrations of the referenced speciesin reactor makeup water should be performed.

3MgO•2SiO •2H O7MgO•8SiO •H O

340 ppm B, pH 8.31

1250 ppm B, pH 7.87

10

10

10

10

10

10

10 10 10 10 10

ppb Mg

ppb

SiO

-1 0 1 2 3

4

3

2

1

0

-1

2

2 2

2 2

Figure 2-11Solubility of Magnesium Silicates at 350°C


2-22

More recent thermodynamic calculations, again using MULTEQ, included examinationof borate precipitates under RCS conditions in the presence of silica and zeolite-formingcations (100). This work concluded that the calcium and magnesium borates precipitatepreferentially over the silicates, except near the end of the cycle when coolant boronconcentrations are reduced. Also, aluminum in hydroxide form was found to remainsoluble and could, therefore, be cleaned up by ion exchange instead of precipitating asa silicate on the core. Thermodynamic models, therefore, suggest that there may in factbe a tolerable risk level of zeolite-forming cation ingress since these species may notimmediately form silicates on the core in borated water.

Silica may be found in high concentrations in the spent fuel and refueling pool watersdue to decomposition of Boraflex. During refueling operations, silica contaminates therefueling water via fuel movement through the transfer canal and from the boric acidstorage tank in those plants that recycle boric acid. Removal of silica can be expensive.Though silicate deposition is a concern, several issues regarding high RCS silica shouldbe noted. The experience of the past 20 years with control of aluminum, calcium, andmagnesium has been excellent, and no unusual events due to silicate formation havebeen reported. In this context, silicon in the range of 2-5% has been found in core cruddeposits at Millstone 3 (44), but the calcium, magnesium, and aluminum concentrationswere low, and no deleterious effects of the silica on crud properties or fuel corrosionwere noted. These data confirm that silica, in the absence of the zeolite-formingelements, is not deleterious to fuel performance. It is also noted that RCS aluminum,calcium, and magnesium are typically low, <5 ppb. Complete precipitation of theelements from these low concentrations will be on the order of several grams ofdeposit. This amount is inconsequential compared to the estimated core crud mass of10-20 kg. Therefore, from a fuel deposit perspective, high coolant concentrations ofsilica are generally not a key technical issue. However, if RCS silica levels are expectedto exceed 1 ppm, maintaining low aluminum, calcium and magnesium concentrationsin the makeup water is considered prudent.

Plant experience with silica >1 ppm has been favorable. Many European and a fewU.S. plants have operated with silica in the range of 1-3 ppm without any reportedproblems. It should be noted that the zeolite forming cationic species in the makeupwater in those high silica cases were all well below the current diagnostic limits.Additionally, a demonstration program is being performed at the Palisades plant toevaluate the effects of operating with up to 3 ppm silica in the coolant on fuel claddingdeposits and oxidation. Fuel surveillance will be performed to confirm acceptability.Although relaxation of the silica limits may be justified based on the existing analysisand plant data base, caution should be exercised in plants with high fuel thermal dutyand known deposit problems, as the data on the effect of silica under such conditionsare limited.

Based on the above considerations, the following recommendations regarding silica inthe RCS can be offered:


2-23

• Silica should be considered a diagnostic parameter during power operation.

• Efforts should be made (e.g., adjusting water levels) during the refueling activitiesto minimize contamination of the reactor water by the spent fuel pool water.

• Chemistry personnel should develop a plant-specific target for RCS silica duringcontinuous power operation, which will be applicable so long as calcium,aluminum, and magnesium in the primary water makeup streams remain withinthe typical ranges experienced at that plant. In writing a technical justification forthis target, consideration should be given to plant-specific issues (e.g., calcium ormagnesium contamination in makeup streams, reactivity concerns during dilutions,tritium discharge in plants recycling boric acid, etc.) and fuel vendor concernsshould be resolved. Experience up to 3,000 ppb has been reported, though cautionshould be exercised in plants with high fuel thermal duty and known crudproblems, as the data on the effect of silica on fuel cladding under such conditions istoo limited to ensure the absence of crud-induced corrosion.

• Chemistry personnel should attempt to reduce silica concentrations below the targetlevel as soon as possible after achieving 100% power operation.

2.4.2.4 Suspended Solids

There currently is no clear correlation between RCS suspended solids concentrationsand deposit magnitudes at operating units. However, early models of depositformation on fuel cladding surfaces included simplistic representations of the effect ofparticulate and solubility driven deposition. In some of these models, a lineardependence on suspended solids concentration was assumed (50, 51). However, it wasgenerally believed that solubility driven deposition governed the overall depositionprocess, and programs for deposit minimization were established based on controllingmagnetite solubility variations across the core.

Within the normal range of suspended solids concentrations (e.g., less than 10 ppb), itis difficult to impose operating limits which suggest an operating response. Normalvariations in suspended solids concentrations are expected despite continuous filtrationvia the cleanup system. However, the occurrence of intermittent particulate releaseshas become more frequent during recent years, and this may be associated withchanges in deposit distribution, magnitude, or composition, all of which are affected bycoolant chemistry as well as by core design. Large, continuous suspended solidsconcentrations (e.g., 350 ppb) may threaten plant operation due to plugging of RCP sealwater filters or damage to RCP seals. Suspended solids may also contribute todifficulties with operation of control rod drive mechanisms.


2-24

2.4.2.5 pH

Within the range of pH employed in the RCS (6.9 to 7.4), pH, by itself, has no effect onfuel corrosion. However, pH has a major effect on fuel crud deposits, which can affectclad corrosion. Also, higher pH is usually associated with higher lithiumconcentrations. (The effect of lithium on corrosion is discussed below.)

Operation below pHTave = 6.9 is not desirable since it is expected to lead to heavier corecrud deposits. At the present time, almost all plants operate at pHTave = 6.9 or above.Operation at pHTave = 6.9 is also recommended at the start of extended fuel cycles, whichmay require boron concentrations in excess of 1200 ppm and lithium concentrations inexcess of 2.2 ppm.

Comparative testing at Trojan and Beaver Valley showed that at a pHT of 6.9 there wassignificantly less crud deposition on the fuel rods than at lower pH (53). When theplant with the lower pH changed to a pHT of 6.9 for the second and subsequent fuelcycles, the new fuel remained clean even though the initial, heavily crudded fuelretained its deposits.

Reference (48) presents additional incidents where unusual crud buildup on fuelsurfaces could be directly correlated to low pH operation. In one instance, low pHoperation led to the formation of localized, dense crud patches. ZrO2 thicknesses underthe crud were up to 110 µm as compared with 25 µm in uncrudded areas of the highpower rods. In another plant, operation at low pH (no lithium additions) yielded crudthicknesses of up to 75 µm in the first operating cycle. ZrO2 + crud thickness was asmuch as 100 µm. The pHTave was controlled at 6.9 during the second cycle and,although crud levels did not significantly diminish on the reinserted assemblies, nosignificant crud was observed on the fresh, Cycle 2 fuel.

2.4.2.6 Lithium

The susceptibility of the Zircaloys to accelerated corrosion in concentrated solutions ofLiOH in the absence of boric acid is well known (54, 55). At a LiOH concentrationcorresponding to 70 ppm Li the corrosion rate is increased by a factor of two to six inwater at 300-360°C (56). A smaller, yet measurable effect was reported at a Liconcentration of 7 ppm (58). Of even greater concern is the acceleration of corrosion inheated crevices, where local subcooled boiling resulted in concentration of LiOH fromnormally benign solutions to those causing pronounced corrosion (57, 58). Opposingthese concerns with LiOH, improvements in fuel economy and dose reductioninitiatives have led to the use of higher lithium levels due to longer cycle lengths.Additionally, higher discharge burnups and fuel thermal duty have eroded corrosionmargins such that, at the current prevailing lead rod discharge burnups of 50-60GWd/MtU, the end-of-life oxide thickness on Zircaloy-4 cladding in many PWRs can


2-25

be expected to reach about 100 µm, which is near the upper limit of acceptability bymost fuel vendors. Recent improvements in Zircaloy-4 cladding material result in 10-30% reduction in the oxide thickness. However, such improvements have not yetyielded enough margin for further burnup extension or for tolerating further increasesin thermal or chemical duties during operation.

As can be seen in Figures 2-2 through 2-4, the integrated exposure of fuel to lithiumthroughout one fuel cycle is increased significantly when the chemistry is changedfrom Coordinated Li/B at pHT 6.9 to Elevated Lithium. An intermediate increase canalso be seen in the case of Modified Chemistry. To evaluate the potential effects of theincreased lithium exposure on fuel cladding corrosion, fuel surveillance programs wereimplemented to obtain cladding oxide thickness data when the Elevated Lithiumregime was implemented at four plants in the late 1980s. Results from St. Lucie-1,Oconee-2, and Millstone-3 have been analyzed by fuel vendors (59-61). In all threecases, it was concluded that the Elevated Lithium chemistry increased the oxidethickness on the order of 10-15%. The certainty of these conclusions, however, must betempered because of large scatter in the data. Comparison of the Millstone-3 data withplant data for Coordinated Chemistry is shown in Figure 2-12.

Figure 2-12Measured Oxide vs. Rod Average Burnup at Millstone and North Anna


2-26

Several recent models that describe the in-reactor corrosion behavior of Zircaloy-4 fuelcladding have been reported (60, 62-65), and these models include contributions of alithium dependent term to the total corrosion. In two of the models (62, 65), thedifference in corrosion between use of the Coordinated and Modified regimes isrelatively small, 10-15%, for fuel near the corrosion limit. The effect of the lithium issomewhat larger in the other model (64). However, another analysis of some of thecorrosion data, including data from two of the Elevated Lithium cycles at Millstone 3,suggests that the observed corrosion is not due to the exposure to the high lithium.This finding, which is in contrast to the aforementioned reports, was based largely ondata from fuel exposed to less than limiting conditions; that is, the burnups were lessthan 50 GWd/MtU and the oxide films less than 80 µm (67). It appears, therefore, thata consensus has not yet been reached within the technical community regarding theeffect of lithium on the in-reactor corrosion of Zircaloy.

At least part of the controversy regarding the effect of lithium on corrosion may lie inthe database analyzed. It is generally agreed that the probability of lithium enhancingZircaloy oxidation increases with oxide thickness (and hence with burnup), and withthe extent of subcooled boiling on the fuel clad surface. This is supported by Frenchstudies showing that lithium penetration into the oxide accelerates during the laterphases of the corrosion process and that the Li enrichment of the oxide increases withvoid fraction (64). It is also clear from laboratory data that the presence of boric acidserves to mitigate the deleterious effect of LiOH (64, 69, 70). With the addition of boricacid in the range of 100-1200 ppm B, the effect of concentrated LiOH in laboratoryautoclave tests is largely reduced; some reported a factor of only two increases at 100ppm Li; others showed no increase at 200 ppm. In general, water containing <10 ppmLi and >100 ppm B as boric acid has negligible effect on Zircaloy corrosion in coupontests.

The effects of Li/B/pH have also been evaluated using electrically heated rods andexposure times of 60 days (64, 69). The results confirmed the coupon test data thatLiOH at a concentration of 10 ppm Li had no or a very small effect on claddingcorrosion with thin oxide (2-3 µm). At a very high void fraction of 35%, the corrosionrate was increased by a factor of ~1.5-2 in water containing 10 ppm Li; no effects werefound at 0 and 5 ppm Li. With 1000 ppm B added to the 10 ppm Li water, only a smallcorrosion enhancement was measured. Overall, the loop test results show that thecorrosion enhancement effect due to Li at <10 ppm is small at low void fractions (<5%).At higher Li concentrations and high void fractions, corrosion enhancement wasdetected. Note that most of the loop tests were performed on fresh cladding, with thinoxide films, usually <10 µm, high Li concentrations, and high void fractions. All ofthese data, and the coupon tests, indicate the increased likelihood of some lithiumenhancement of corrosion for thick oxide films with subcooled boiling, or with boilingin the pores of thick oxides. Thus, the enhancement will be sensitive to specific powerhistories, with a possible coupling of the local thermal duty to the Li/B ratio


2-27

throughout life. Therefore, the correlation for prediction of the Li effect on fuelcladding corrosion in PWRs requires calibration to actual fuel rod corrosion data fromnuclear power plants. The several corrosion models calibrated to plant data suggestless than a 15% effect.

Extracting fuel rod corrosion data for assessment of the lithium effect is rathercomplicated, as there are many factors known to contribute to Zircaloy in-reactorcorrosion. One of the factors that influences fuel cladding corrosion is the extent ofcorrosion product deposits on the fuel. Deposition on the fuel has generally beenunder control and of little consequence to fuel performance since the introduction of thecoordinated Li/B coolant control strategy in the early 1980s. Further control of corecrud deposits as evidenced by the reduction of out of core fields was effected by the useof the higher pH scheme of the modified Li/B strategy that was implemented in plantsbeginning in the mid 1980s. Simultaneous with these actions to reduce systemcorrosion and fuel rod crud deposits, the thermal duty on the fuel was steadilyincreasing, so subcooled nucleate boiling was becoming responsible for larger fractionsof the total heat transferred to the coolant. Subcooled boiling exacerbates cruddeposition as demonstrated in both loop tests (71) and plant data (44, 72). (Thecombination of subcooled boiling and crud deposits is the cause of the Axial OffsetAnomaly, which is described in Section 2.4.4). Crud deposits contribute directly to fuelrod corrosion by providing an additional barrier to heat flow (45, 65), although theeffect is usually not great because the deposits are thin and highly porous. However,with the subcooled boiling duty currently experienced on advanced fuel in hightemperature plants, the effect of the crud on corrosion can become similar to the effectof Li between the coordinated and modified Li/B strategies for high-power two-cyclefuel (65). Thus, for fuel performance the different coolant strategies betweencoordinated and modified may have little effect on cladding corrosion when the effectof coolant chemistry on crud deposition is taken into account. Similar arguments mayapply to higher pH and Li strategies, but it is yet to be demonstrated that suchstrategies, by themselves, are effective in reducing crud under the subcooled boilingexperienced by today’s high efficiency fuel designs.

2.4.2.7 Zinc

A cooperative program was undertaken by EPRI and the Westinghouse Owners Groupto evaluate the effectiveness of zinc in mitigating Alloy 600 PWSCC and radiationfields, and to assess the effect of zinc on corrosion of fuel cladding. Effects of zinc onPWSCC and radiation fields are discussed in Sections 2.3 and 2.5, respectively, andeffects on fuel are discussed here. The program included evaluation of zinc addition atFarley-2 and the Halden test reactor. The scope of the program at Farley-2 includedpre- and post- zinc measurements of fuel cladding oxide growth, fuel crud analysis aswell as radiation dose rates, PWSCC of steam generator tubes, and chemistry andradiochemistry monitoring during Cycle 10 (38, 39).


2-28

Prior to zinc addition at Farley, EPRI conducted an in-pile corrosion test program at theHalden test reactor simulating PWR conditions. The objective of this project was toprovide early warning of any potential substantial adverse effects of zinc on fuelcladding corrosion. At Halden, several cladding segments were exposed to 50 ppb zinc(added as zinc formate) for three irradiation cycles of approximately 100 days each (73).The overall conclusion from these tests was that there was no unusual growth ofzirconium oxide in either fresh or pre-irradiated segments after exposure to zinccontaining coolant. Hot cell examinations of two rodlets showed that the averagehydrogen concentrations were normal and within expectations. Halden resultsindicated that zinc had no substantial adverse effect on corrosion and hydriding of fuelcladding.

At Farley-2, zinc was added as natural zinc acetate after 190 days into Cycle 10. Thecoolant zinc concentration was maintained in the range of 35-45 ppb with a target valueof 40 ppb for approximately 9 months. At the end of Cycle 10, evaluations were madeof cladding oxide growth and crud scrapings. Visual examination of fuel at the end ofCycle 10 showed a uniform black deposit over fuel assemblies including bottomnozzles and grids. Crud scrapings from upper spans of fuel rods showed that thedeposit was thin (less than one micron). Chemical analysis of the deposit suggests thatit consisted of nickel ferrite with excess nickel, indicating the likely presence of nickelor nickel oxide. The amount of zinc in the deposit was small, less than 5%. Eddycurrent examination of fuel rods at the end of Cycle 10 showed greater than expectedzirconium oxide thickness compared to measurements made at the end of Cycle 9.Based on initial eddy current data, twice-burned fuel assemblies were removed fromthe core and zinc addition was suspended during Cycle 11 pending further evaluationof the observations. During this evaluation, data from Farley-1 Cycle 13 (non-zinc)were also reviewed. Fuel oxide thickness in both Farley-1 Cycle 13 and Farley-2 Cycle10 was higher than that in the pre-zinc Cycle 9 at Farley-2. The conclusions of the rootcause evaluations were that differences between the observed and the expected cladoxidation behavior must have been caused principally by fuel duty and inherentvariability in the cladding corrosion process. Primary coolant chemistry was a minorcontributor. Zinc addition had no more than a small detrimental effect on theobserved fuel clad oxidation.

No zinc was added during Cycle 11 at Farley-2. Fuel thermal duty was lower in Cycle11 and this resulted in lower cladding oxidation consistent with predictions of the EPRIPWR Fuel Cladding Corrosion Model. Farley-2 resumed zinc addition in Cycle 12.However, the period of zinc addition during Cycle 12 was too short (3 months) toassess its effect on fuel cladding corrosion.

Diablo Canyon Unit 1 commenced adding natural zinc in June 1998, and plans to add itfor approximately eight months. The effect of zinc on fuel cladding corrosion will bemonitored under an EPRI-PG&E sponsored program.


2-29

Zinc addition has also been implemented in Biblis, Unit B in Germany which isoperated by RWE Energie AG (74). The objective of the program is to reduce shutdowndose rates and postpone the need for full system decontamination. Depleted zincacetate was added in Biblis Unit B during Cycle 17. Zinc addition was initiated in earlySeptember 1996 and zinc was detected in the coolant in mid-October 1996 after adosage of approximately 1 kg of zinc. The zinc concentration was ~ 5 ppb and 3.5 kgof zinc was injected during the cycle. There was no adverse effect of zinc on fuelcladding oxidation. The concentration of zinc in the fuel crud was very low. Zincaddition will be continued during the next cycle. In early 1998 two other GermanPWRs, Biblis A and Obrigheim, commenced zinc addition.

With regard to assessing zinc for plants operating at elevated silica levels, there areonly limited thermodynamic data on zinc silicate precipitates. Whether or not zincsilicates would become a fuel concern under these conditions is not known at this time,but is under investigation.

2.4.3 Alternate Cladding Material to Zircaloy-4

With realization that corrosion limits are being approached due to the fuel dutiesimposed by economic requirements, appreciable activity is under way on thedevelopment of new and improved zirconium-based alloys for fuel cladding andstructural applications. At least three of these materials have been approved bylicensing authorities and are in use on full-region bases; ZIRLO in the U.S. and the ELSDuplex Clad and the M5 alloy in some European countries. Additionally, all vendorsare reporting in-reactor results from various cladding alloy demonstration programs.As these alloys are screened extensively in autoclave tests, including resistance toconcentrated LiOH, they offer the promise of greater flexibility in the allowable rangesof pH and Li than exists for Zircaloy-4. Performance of these new alloys requires closesurveillance under a variety of operational duties and coolant chemistries to ensuretheir acceptance as robust materials in the future.

2.4.4 Effect of pH on Fuel Axial Offset

The Axial Offset Anomaly (AOA) results from a flux depression in the top half of thecore such that power is skewed to the bottom. The flux depression is caused by theconcentration of soluble boron in porous crud deposits as a result of subcooled boiling(44). It is believed that boron concentrates to the extent that a boron compound, mostlikely LiBO2, precipitates within the crud deposit. It is the precipitated boron-rich solidthat is responsible for the flux depressions observed. There are three conditions thatmust be present for AOA to occur: soluble boron, a layer of porous crud of sufficientthickness to serve as the medium for concentration of the soluble boron, and sufficientheat flux to cause subcooled boiling at the surface. In PWRs with high coolant


2-30

temperatures, these conditions exist for a large fraction of the feed fuel, and it is thesehigh power fuel assemblies that drive AOA.

The association of the extent of crud deposits with the occurrence of AOA has beenestablished (44, 72), and the deposits are noted to be heavy on feed fuel. Theoccurrence of AOA at burnups of 4000 to about 8000 MWd/MtU indicates thatsufficient crud is deposited in the first 4-8 months of the cycle to cause the problem. Itis also noted that AOA has been observed in plants and cycles using both Coordinatedand Modified Li/B control strategies. However, not all high temperature plants/cycleshave experienced AOA, indicating that a complete explanation for this phenomenonhas not yet been developed. Attempts have been made to relate corrosion productrelease rates (76) and concentrations in the coolant (44) to AOA, but no unequivocaldata have been obtained.

In an attempt to minimize crud deposition early in the cycle, several plants arecurrently operating, or considering operating, with a pHT of 7.0-7.2 throughout thecycle. At these pHs the beginning-of-life lithium will be in the range of 3.0-3.5 ppm foran 18-month cycle and the lithium may be above 2.2 ppm for as much as half the cycledepending upon the core design. This form of Coordinated Chemistry has beenproposed (75) and is being implemented with vendor approval as necessary. It isgenerally believed that the risks associated with any additional corrosion due to thehigher lithium in the first half of the cycle will be more than compensated by thepotential for decreased crud deposition. Hopefully, fuel and dose rate surveillanceprograms will be performed upon completion of some of these cycles to verify andquantify the effects and form the bases for future decisions on use of high pH. Fuelsurveillance is particularly desirable if a high pH and high lithium operation isimplemented in plants with significant subcooled boiling. In the event that thebeginning-of-cycle lithium exceeds the experience base of 3.5 ppm, a fuel surveillanceprogram is recommended.

It should be noted that some plants have had good success with and still use thecoordinated pHT 6.9 lithium/boron control strategy. The advantages are minimizationof possible Li effects on Zircaloy corrosion and Alloy 600 cracking. The disadvantagesare in possible acceleration of crud deposition on the fuel and transport of activatedspecies to out of core areas. If plant experience has been favorable, or margin exists toaccommodate these potential effects, the pHT 6.9 coordinated chemistry remains aviable option.

Although the basic mechanism of AOA is understood, the role of coolant chemistry inthe phenomenon is an active area of research that is included in the EPRI Robust FuelProgram (77). Tasks being pursued relative to water chemistry issues include: 1) aChemistry Diagnostics effort to more fully understand the relationship between coolantchemistry, crud transport and deposition on fuel, 2) a loop test program aimed atdeveloping a more quantitative relationship between subcooled boiling and crud


2-31

deposition, and 3) a thermal and hydraulic analysis activity to establish a methodologyfor the analytic description (78) of subcooled boiling at fuel surfaces and to developcriteria for acceptable boiling duty without AOA. Results of these activities may leadto development of criteria that will limit the efficiency of core designs in hightemperature plants until a reliable means of controlling crud in the presence ofsubcooled boiling can be developed.

With our understanding of the effects of pH on crud deposition in general, butacknowledging the uncertainty of the quantitative role of subcooled boiling on cruddeposition, the following recommendations are offered for minimizing the risk of AOAfor plants with high boiling duty:

• Minimize crud deposition on fuel by increasing the beginning-of-cycle pH to thehighest level allowable within the range of 6.9-7.2 and for a beginning-of-cyclelithium limit of 3.5 ppm. If the starting lithium exceeds 3.5 ppm, fuel corrosionconcerns should be addressed and a fuel surveillance program should beconsidered.

• For the Li/B strategy chosen, control the lithium concentration within tight limits, atleast as tight as those given in Table 3-1.

• Maximize corrosion product removal from the RCS during shutdown and startupoperations while minimizing the impacts on refueling critical path times andachievement of full power, respectively.

2.5 Radiation Field Control

2.5.1 Sources of Radiation Fields

Shutdown radiation fields can be affected by numerous factors including operationalchemistry control, plant configuration, materials of construction, surface finish, andtime of operation (53, 79-82).

The primary long-term source of radiation fields in PWRs is cobalt-60, formed byneutron capture in cobalt (cobalt-59). Cobalt-58, formed by the nickel-58 (n,p) reactionwith fast neutrons, is the second most important contributor to out-of-core radiationfields. The major source of nickel in the reactor coolant arises from the corrosion of theAlloy 600 steam generator tubing. The major sources of cobalt are the corrosion releaseof cobalt impurity in the Alloy 600 or 690 tubing (83), mechanical wear of high-cobaltalloys (used in control rod drive mechanisms, reactor coolant pumps, valves, etc.), andcorrosion release of cobalt from stainless steel piping and vessel internals. Cobalt andnickel are released to the primary coolant in the corrosion process and are deposited onfuel surfaces. The activation products, Co-58 and Co-60, are gradually released to and


2-32

circulated by the coolant in soluble, insoluble, or colloidal forms. Although partiallyremoved by the purification system, incorporation of these species into the corrosionfilms on the system surfaces results in the buildup of radiation fields.

Cobalt-58 is the primary contributor to radiation fields in early plant life and cobalt-60becomes predominant in later life. As shown in Figure 2-13, typical radiation levelsreach a peak between 4 and 6 EFPY and then level off or decline. The fact thatradiation levels vary from plant to plant indicates that factors other than operating time(e.g., coolant chemistry, materials of construction, and surface finish) have an effect onradiation fields (84). Plant operating history, i.e., power maneuvers, number of start-ups and shutdowns, etc., also may have an effect (85).

Range of PlantMeasurements

1-Year Moving Mean

100

10

10 2 4 6 8 10 12 14 16

Effective Full Power Years

Dos

e R

ate

(rem

/hr)

Figure 2-13Radiation Field Trends in PWRs


2-33


Accumulation of activated corrosion products on out-of-core surfaces is the main causefor buildup of radiation fields that persist after shutdown. There are three essentialsteps in this process. First, corrosion products are released to the coolant from out-of-core surfaces. Second, some of this material is deposited in the core region, where it isactivated by exposure to a neutron flux. Third, some of the activated material isreleased from the core and incorporated in out-of-core oxides. Primary coolantchemistry influences each of these steps, and its variation during the fuel cycle has animportant effect on the accumulated amount of out-of-core radioactivity.

Primary coolant pH is the key parameter in coolant chemistry. Corrosion rates of out-of-core materials, corrosion product release rates from these materials, solubility ofcorrosion products, the properties of corrosion product particles suspended in thecoolant, the thermodynamics and kinetics of in-core deposition, and release andredeposition out-of-core are all partially related to pH. The pH in the primary circuit isdependent on the local temperature and the existing water chemistry, primarily theconcentrations of lithium and boron.

Corrosion products are carried by primary coolant in two forms: as solids (i.e., metaloxide particles) and as dissolved species such as metal ions and their complexes withhydroxyl ions. Corrosion product deposits are metal oxides. Deposits on fuel in thecore contain nickel ferrites with a general formula given by NixFe3-xO4, nickel oxide, andpossibly nickel metal. If account is taken of the low (but operationally important)cobalt content, the general formula is CoyNixFe3-x-yO4, in which y is much smaller than x.Films on the interior of steam generator tubing have an inner layer that is a nickel-chromium-iron spinel and an outer layer that is nickel ferrite. The chemicalcomposition of nickel ferrites may differ from place to place, since temperature andcoolant chemistry conditions vary around the primary loop, determining whichcompositions are stable at any given location.

More recently, samples of fuel deposits from plants operating thermally uprated cores(some of which have experienced axial offset anomaly) have been shown to contain anexcess of nickel compared to that expected from typical nickel ferrite experience. Infact, fuel scrapes from Callaway Cycle 6 (an AOA cycle) demonstrated x-ray diffractiondata which would place the nickel lattice parameter in the ferrite closer to x = 0.3 thanbetween 0.5 and 1.0 (44). The balance of nickel is reasoned to be either in a metallic or amono-oxide form based upon coolant samples. Samples of deposit flakes obtained atthe end of Cycle 9 were composed primarily of nickel oxide crystals (102).

The overall process of activity accumulation on out-of-core surfaces involves a complexset of chemical and physical mechanisms interacting with each other through themedium of the primary coolant and its chemistry. Nevertheless, the indicators for thethree important factors all point in the same direction: namely, that operation at coolant


2-34

pH in the range from 7 to 7.4 at 300°C can be expected to minimize radiation fieldbuildup, with best results at the high end of the range. Each of these three factors isdiscussed below.

2.5.2.1 Corrosion Product Release

Corrosion products containing iron, nickel and cobalt enter the primary coolant mainlyas a result of general corrosion of out-of-core surfaces. This form of corrosion proceedsin primary coolant at such a low rate that it has no measurable effect on the integrity orperformance of components, but it is of concern because of its consequences foractivation and radioactivity redistribution and because it is the source of fuel depositswhich can affect fuel performance, as discussed in Section 2.4.

It has long been recognized that general corrosion of carbon steel at temperaturescorresponding to primary coolant conditions is at a minimum in otherwise pure waterwhen the pH is raised by a strong base to around 10 to 11, measured at roomtemperature (86). This corresponds to pH from 7.4-8.4 at 300°C. General corrosion ofstainless steels and Alloy 600 behaves similarly. Furthermore, laboratory tests indicatethat corrosion product release rates for both Alloy 600 and stainless steel are low incoolant with pH300°C = 7.0 and even lower at pH300°C = 7.4 (87-90).

2.5.2.2 Particulate Transport

Deposits on fuel and corrosion product particles in primary coolant are visible evidenceof corrosion product migration. Particulates in the coolant have received muchattention, but quantitative treatment of the processes by which they are deposited onand released from in-core and out-of-core surfaces has been difficult. Nonetheless, theradiation management impact of uncontrolled particulate release from the core duringa shutdown transient is demonstrated by two recent plant experiences wherein highspecific activity, highly nickel-rich deposits were discharged into the coolant duringcooldown (101). Although no definitive root cause has been established, both plantshad operated with thermally uprated cores and neither can exclude a combination ofthermal hydraulic and chemistry disturbances as the cause for the releases.

Any attempt to understand the behavior in primary coolant of colloids, and largerparticles as well, must consider the effects of the surface density of electrical chargesthat metal oxides acquire when immersed in aqueous solutions. Electrical chargesaffect the tendency of particles to deposit on another solid surface or to agglomeratewith each other. The net charge on metal oxide surfaces is influenced strongly by pH.


2-35

2.5.2.3 Soluble Transport

Dissolved corrosion products can be transported, dissolved or deposited by the coolantbased on solubility differences. The solubilities of nickel and iron depend ontemperature, pH and redox potential, all of which vary with position around the non-isothermal loop. Also, the pH at each location changes with time as boron is removedduring the fuel cycle, and redox potential may sometimes be altered if there is a changein dissolved hydrogen concentration. These differences cause deposits to grow byprecipitation when concentrations of dissolved corrosion products exceed solubilitylimits and to dissolve when concentrations are lower.

Correlating solubility as a function of temperature, pH and redox potential, allows oneto examine the consequences of alternative strategies for control of lithiumconcentration as boron concentration is decreased. A key result from initial correlationof nickel ferrite solubility (84) was that the temperature coefficient at 300°C forsolubility of iron (assuming constant concentration of lithium, boron and hydrogen)was predicted to change sign (from retrograde solubility to "normal" temperaturedependence) at a pH300°C of about 7.4,* which was supported by later correlations (91,92).

There are several computer codes that address the radiation field buildup problem bysimulation of processes in the primary system. Most of these embody a dozen or moremechanistic models of specific mass transfer processes, with empirical or estimatedcoefficients. More appropriate for the present purpose is CRUDSIM (84), aconceptually simple model that addresses corrosion product and activity transportresulting from precipitation/dissolution processes. Solubility differences for nickelferrites are the only driving forces for transport considered in this model. Calculatedsolubility for iron from nickel ferrite and for nickel from nickel oxide are given inFigures 2-14 and 2-15 as a function of pH and temperature (76). Although CRUDSIMemploys many simplifying assumptions, it has been reasonably successful in assessingthe relative effects of various coolant chemistry conditions on radiation field buildup.

* This is a distinctly more alkaline condition than the pH of 6.8 for zero temperature coefficient that is indicated bythe correlation developed by Sweeton and Baes for their own data on magnetite solubility. The discrepancy isprobably because their data did not provide an accurate estimate for the third hydrolysis constant of ferrous ion,which is important in alkaline water. Their measurements were mainly done with solutions on the acid side ofneutral.


2-36

Core Inlet Core Outlet

Iron S

olubil

ity, p

pb

Temperature, C

pH = 7.5

pH = 7.4

pH = 7.3

pH = 7.1

pH = 6.9

270 280 290 300 310 320 330 340 3503.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

Tavg

Tavg

Tavg

Tavg

Tavg

Figure 2-14Variations in Iron Solubility from Core Inlet to Outlet as a Function of pHT at theCore Average Coolant Temperature (Boron = 600 ppm, H2 = 35 cc/kg H2O)

x

Nickel

Solub

ility, pp

b

Temperature, C270 280 290 300 310 320 330 340 350

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

Core Inlet Core Outlet

x

x

x

x

x

xx

x

pH = 6.9

pH = 7.1

pH = 7.3

pH = 7.4

pH = 7.5

Tavg

Tavg

Tavg

Tavg

Tavg

Figure 2-15Variations in Nickel Solubility from Core Inlet to Outlet as a Function of pHT at theCore Average Coolant Temperature (B = 600 ppm, H2 = 35 cc/kg H2O)


2-37

Among other things, pHT varies with hot leg and cold leg temperatures. By operatingwith pHTave = 7.4, the solubility of iron in equilibrium with nickel ferrite shoulddecrease slightly in the core inlet portion (slightly depositing) and increase slightly inthe core outlet (slightly dissolving). One way to evaluate the effects of alternativelithium control strategies on radiation field buildup by solution transport processes isby computer simulation with plant-specific parameters. It is not likely that any singlerecommendation on pH will give best results for all or even most plants. However,there is no reason to doubt that the contribution of solution transport to out-of-pileradioactivity inventory can be minimized by coolant chemistries for which pH at 300°Capproaches 7.4 to 7.5 as early in the cycle as possible.

2.5.2.4 Verification Tests and Plant Experience

The preceding sections have discussed three important factors that affect radioactivityaccumulation on out-of-core piping and components due to activation of deposits onthe core: corrosion product release, particulate transport and solution transport. Thediscussions indicate that activity transport effects of all three factors should be reducedby coolant chemistry with high temperature pH between 7 and 7.4, with most effectivemitigation at the higher end of the range. Uncontrolled releases of particulates fromhigh reactivity regions of the core have occurred during changes in pH, temperatureand thermal hydraulics at or near shutdown (101).

Extensive tests with pressurized in-pile loops have confirmed that ex-core radioactivityis minimized by increased coolant pH (93, 94). The MIT loops were designed to modelthe primary circuit of a representative, non-boiling PWR plant. Seven one-month-longin-pile runs were carried out at lithium and boron concentrations selected to givepH300°C = 6.5 to 7.5. The results are summarized by Figure 2-16, which shows the effect ofpH on activity per unit area of the steam generator sections of the loops after the in-pileruns. Activities are normalized by megawatt-hours to allow results for two longer runs(indicated on the plot by A2 and B2) to be compared with the shorter runs. There issignificant decrease of surface activity for both Co-60 and Co-58 as pH increases.Although this decrease may be caused by combined effects of pH on corrosion releaseand on both particulate and solution modes for corrosion product transport, the MITteam concluded that their results were consistent with the view that metal oxidesolubility plays a dominant role (93).


2-38

Figure 2-16Variation with pH of Average Activities of Co-60 and Co-58 on Steam GeneratorTubing at the End of MIT in-Pile Loop Tests

2.5.2.5 Plant Data

Since operation with various coolant chemistry regimes would be expected to influenceplant radiation fields, a number of evaluations of plant dose rate data and special planttests have been performed since 1980 in an attempt to define and/or corroborate thechemistry effect. However, other factors such as variations in cobalt input, componentsurface treatments, core thermal design and unusual releases of crud can also affectdose rates. The effects of these factors have been taken into account in the evaluationswhen information was available. This section summarizes the results of historicalstudies designed to evaluate the effects of the various coolant chemistry regimes onplant dose rates.

Uncoordinated versus Coordinated Chemistry Operation

The first evaluations in 1984-1985 of the effects of pH on plant radiation fields indicatedthat operation with a lower pH as compared to a higher pH or coordinated pH 6.9coolant chemistry resulted in generally higher dose rates (84, 95). The evaluations usedsteam generator channel head dose rate data from about thirteen plants from the firstone or two cycles of operation. The evaluations were based on a quality factorquantifying the degree of adherence of the operational chemistry relative tocoordinated pHT 6.9.

A later evaluation in 1992 (96) was based on data from about thirty-six plants and anumber of cycles of operation and used two evaluation techniques (arithmetic averageof all the data and a 1-year moving mean trend) and dose rate data representing the


2-39

steam generator tubing and crossover piping in addition to the steam generator channelhead. The evaluation concluded that operation with Coordinated Chemistry comparedto uncoordinated chemistry resulted in a reduction in dose rates of about 18%.

More recently, EPRI has suggested considering operation at constant pHT > 6.9 , such asconstant 7.1, but within fuel and material integrity constraints reviewed earlier inSection 2 in order to reduce the release rate of soluble fuel deposit precursor speciesfrom the steam system surfaces (75). This strategy is expected to be most attractive toplants more concerned with total core crud deposit buildup (such as those with coressusceptible to anomalous axial flux depressions) than with optimizing soluble activitytransport (current lithium limits preclude end of cycle operation at pH 7.4 when pH iselevated early in the cycle). Currently, several utilities are employing some version ofconstant but elevated pH during eighteen month cycles; shutdown dose rate data willbe available beginning in middle to late 1999. Section 4 provides a suggestedprocedure for making plant-specific decisions for optimizing the pH program.

Modified versus Coordinated Chemistry Operation

The first evaluation of the effect of initial operation with Modified Chemistry operationwas made in 1992 (97). The evaluation was based on comparison of dose rates fromfour plants that operated since startup with Modified Chemistry and ten other plantswith similar characteristics that started up with Coordinated Chemistry. Comparisonof the dose rates following two cycles of operation between the two groups of plants atthree locations indicated that initial operation with Modified Chemistry resulted indose rates about 25% lower compared to operation with Coordinated Chemistry. Thecomparison was made by computing the average dose rate at the various locations atthe average EFPY for the two groups of plants. No statistical evaluation of the datawas made.

A second evaluation of Modified versus Coordinated Chemistry operation was made in1994 (98). This evaluation was based on operation for two cycles and utilized data fromthe same four Modified Chemistry plants and three of the ten Coordinated Chemistryplants as used in the evaluation noted above. However, data from two other pH 6.9plants not included in the above study were included. In addition, the coolant pH wascalculated based on the plant chemistry data. (In the first evaluation, the coolantchemistry was assumed to follow the various regimes based on data provided or verbaldiscussion with plant personnel). The evaluation was made using two techniques:averaging the data after two cycles of operation, and comparison of the slopes of thelinear regression of the dose rates versus cycle average pH values for one cycle anddose rate changes between one and two cycle operation as a function of changes in theaverage pH. The evaluation concluded that operation with Modified Chemistryyielded an average reduction in dose rates of about 15% relative to Coordinated


2-40

Chemistry operation. The confidence level of the chemistry effect varied from about 70- 98% depending upon the location.

Elevated Lithium versus Coordinated Chemistry Operation

A number of plants (Ringhals 2, 3, and 4) have operated under Elevated Lithiumcontrol for several cycles and an Elevated Lithium test was run at Millstone 3 for twocycles following one cycle of Coordinated Chemistry (99). Evaluation of the results wasperformed by comparing the dose rate trends at three locations in the Elevated Lithiumplants to a reference group of plants operating with Coordinated Chemistry andcomparing the dose rate trends of the plants to that predicted by the CORA crudtransport code. It was concluded that operation under an Elevated Lithium regimeafter one cycle of Coordinated Chemistry resulted in a reduction of dose rates of about15% compared to operation with Coordinated Chemistry.

A second evaluation of the Elevated Lithium effect was performed by comparing cycle1 and 2 dose rates trends at three locations in the five plants operating withCoordinated Chemistry noted above. The results showed an estimated dose ratebenefit of about 25% (98).

Comparison of Radiation Field Trends as a Function of Coolant Chemistry

Figure 2-17 illustrates the overall trends in plants with Coordinated, Modified andElevated Lithium chemistries using the steam generator channel head radiation fieldsas an index. The data are from a group of plants which have the same factors (otherthan coolant chemistry) that are expected to affect radiation fields (96). Using the three-year moving mean as an index, the radiation fields in the four plants that started upwith Modified Chemistry are about 20% lower than those in the twelve plants thatstarted up with Coordinated Chemistry. Note that the radiation field trends in the twoplants that operated for two or more cycles with Elevated pH 7.4 operation (Millstone 3and Ringhals 4) level off at a lower value after operation with the Elevated pH. In bothplants, reverting to a lower pH resulted in an increase in the radiation fields for one ortwo cycles. The large increase in Millstone-3 is also partially attributed to increasedcrud deposition on the fuel due to subcooled boiling. The trend was reversed inMillstone-3 after using Modified Chemistry.

The trends in Figure 2-17 indicate the average radiation field benefit of ModifiedChemistry operation, but operation under Elevated Lithium would provide a largerbenefit. However, given the material corrosion concerns regarding operation above 2.2ppm lithium for a significant amount of time, a plant-specific optimization programgeared toward determination of the best overall pH program from among optionsconsistent with the principles of Section 3 should be undertaken (see Section 4).Operation with enriched boric acid (EBA), summarized in Appendix C, would alleviatethe material corrosion concerns and provide an increased radiation field benefit.


2-41

Figure 2-17Steam Generator Channel Head Dose Rates (Comparison of Trends of Plants inGroup 3 - Plant Startup after 1981) Data for Four Plants are Shown as Three-YearMoving Mean

2.5.2.6 Zinc

Previous work has shown that low concentrations of zinc in the PWR primary coolantcould reduce out-of -core exposure rates and primary water stress corrosion cracking(PWSCC) of Alloy 600. A cooperative program was undertaken by EPRI and theWestinghouse Owners Group to evaluate the effectiveness of zinc in mitigatingradiation fields and also to determine the effects of zinc additions on Alloy 600 PWSCCand corrosion of fuel cladding. The program included pre- and post- zincmeasurements of radiation dose rates at Farley 2 as well as chemistry andradiochemistry monitoring during Cycle 10 (38, 39). Zinc was added as natural zincacetate after 190 days into Cycle 10. The coolant zinc concentration was maintained inthe range of 35-45 ppb for approximately 9 months. At the end of Cycle 10, evaluationswere made of radiation dose rates as well as fuel cladding oxide growth, crudscrapings, and PWSCC of steam generator tubes. Dose rate effects are discussed below,while effects on PWSCC and fuel are discussed in Sections 2.3 and 2.4.


2-42

Measurements of dose rates after Cycle 10 at Farley 2 with zinc addition showed areduction of 24% at steam generator channel heads of which 11% was attributed to zincaddition and shutdown chemistry practice. Zinc-65 was less than 10% of theradioisotopic mix and only a minor contributor to the radiation fields. A dose saving of40 man rem per fuel cycle was estimated for Farley 2 after five cycles of natural zincaddition.

No zinc was added during Cycle 11 at Farley-2. Dose rate measurements at the end ofCycle 11 showed that the downward trend in dose rates observed from non-zinc Cycle9 to zinc Cycle 10 was reversed at end of Cycle 11 (39). Farley-2 resumed zinc additionin Cycle 12. However the period of zinc addition during Cycle 12 was too short (3months) to assess its effect on dose rates, fuel cladding corrosion and PWSCC.

Diablo Canyon Unit 1 commenced adding natural zinc in June 1998, and plans to add itfor approximately eight months. The effect of zinc on radiation dose rates, fuelcladding corrosion, and PWSCC will be monitored under an EPRI-PGE sponsoredprogram.

Zinc addition has also been implemented in Biblis, Unit B in Germany which isoperated by RWE Energie AG (74). The objective of the program was to reduceshutdown dose rates and postpone the need for full system decontamination. An in-plant surveillance program was conducted which consisted of chemical andradiochemical analysis of coolant, dose rate and gamma activity measurements, fuelinspections, examination of reactor coolant pump seals and examination of oxide layerson steam generator manway cover seals. Depleted zinc acetate was added in BiblisUnit B during Cycle 17. Zinc addition was initiated in early September 1996 and zincwas detected in the coolant in mid-October 1996 after a dosage of approximately 1 kg ofzinc. The zinc concentration was ~ 5 ppb and 3.5 kg of zinc was injected during thecycle. Dose rate measurements at the end of the cycle showed an average dose ratereduction of ~12%. There was no adverse effect of zinc on fuel cladding oxidation. Theconcentration of zinc in the fuel crud was very low. Zinc addition will be continuedduring the next cycle. In early 1998 two other German PWRs, Biblis A and Obrigheim,commenced zinc addition.

2.5.3 Summary

The current trend toward longer fuel cycles has placed an added concern onoptimization of RCS chemistry. Despite application of an optimum chemistry controlprogram, higher radiation fields may be observed. The reduced frequency of outage-related work, proper work planning and application of shutdown chemistry controlsmay decrease the impact of longer fuel cycles.


2-43

Based on laboratory and loop studies it is expected that operation under ElevatedLithium or Modified Chemistry regimes should result in lower dose rates and less crudon the fuel compared to operation with Coordinated Chemistry. Evaluation of plantdata has generally confirmed this expectation in that operation with ModifiedChemistry results in dose rates about 20% lower than with Coordinated Chemistry.Modified Chemistry, however, may not be optimal in those plants that have fueldesigns with subcooled boiling, such as plants susceptible to AOA. In these cases,operation at elevated, constant pH early in the cycle to reduce the source term for cruddeposition on the core may be preferable from a core performance viewpoint. Theeffect of such a protocol on radiation fields ex-core will not be fully known untilanalysis of future shutdown data is obtained on the few plants currently using aconstant, elevated pH program.

Zinc addition to the primary coolant appears promising with regard to reducingradiation dose rates, and is being used at several plants. However, effects on fuelcladding and PWSCC also need to be considered when evaluating possible use of zinc(see Sections 2.3 and 2.4).

3-1

3 POWER OPERATION CHEMISTRY CONTROL

RECOMENDATIONS

3.1 Introduction

In most cases, maintaining chemistry parameters below a certain threshold value orwithin a certain range can minimize negative impacts of chemistry on materialsintegrity, fuel rod corrosion and radiation fields. Each plant must consider the degreeto which it may be subject to potential negative impacts in each of these areas, anddesign their chemistry program appropriately to achieve optimal performance.Recommendations on how to accomplish such an optimization process are presented inSection 4.

In this section, chemistry control recommendations intended to serve as industry-consensus guidelines are presented. These recommendations are based on currenttechnical understanding and industry experience. For site-specific modifications tothese Guidelines, plant chemistry personnel must produce a written technicaljustification that addresses the relevant regulatory, fuel warranty, and safetyimplications of such modifications (see Section 4).

3.2 Reactor Coolant System (RCS) pH Optimization

3.2.1 Operating Recommendations

As noted in Section 2, the optimum approach to pHT control relative to minimizingplant dose rates could be detrimental to materials integrity or fuel corrosion. Likewise,an operating regime that may be optimal for minimization of lithium-related fuelcorrosion might increase corrosion product deposition on fuel and actually enhancecorrosion, or promote axial flux depressions in cores with substantial subcooledboiling. Figure 3-1 attempts to convey the generic problem associated withrecommending an optimum operating pH. This figure is a schematic presentation ofthe concerns discussed in Section 2; no quantitative effects should be inferred. Notethat this figure depicts a typical operating range reflecting the technical bases discussed

Power Operation Chemistry Control Recomendations

3-2

in Section 2, where both materials and fuel integrity are maximized at the expense ofslightly higher radiation fields.

There are also additional issues specific to plants with a history of AOA that maydevelop optimized chemistry programs that emphasize minimization of corrosionproduct deposition on the core. Thermal hydraulic guidelines for assessing the AOArisk of core designs are being developed by the EPRI Robust Fuel Program (GC-110069,“PWR Axial Offset Anomaly (AOA) Guidelines-for final review,” December 1998) (seealso 104).

Reducing the available research and field data associated with RCS pH and impuritiesand their impacts on materials integrity, fuels corrosion, and radiation control into a setof recommendations is not a simple process, e.g., much of the corrosion data areextrapolated from short-term laboratory studies. In addition, plant-specific variables,such as temperature, PWSCC susceptibility, and fuel deposit concerns makeoptimization of pH control a plant-specific determination. Plants should weigh relativerisks of fuel deposit build-up, fuel cladding and system material corrosion, and out-of-core radiation field concerns at each site. While one pH program will not solve allproblems, balancing the benefits and risks will limit detriments to the primary system.Optimization methods are discussed in detail for material integrity and crud transportissues in Section 4, and in Section 2.4 for AOA concerns. Also, a set of "Principles" isgiven in Table 3-1 and is recommended for use in developing a plant-specificboron/lithium control scheme.

Baseline guidance for implementing these Principles is depicted schematically inFigures 3-2(a), showing risk to fuel integrity, and 3-2(b), showing PWSCC risk to Alloy600.

• “Red” is meant to denote substantial downside risk either because of knowndetriments to fuel integrity (e.g., Principle 1) or because industry experience isextremely limited. As noted in Section 2, the industry experience base is such thatthe lithium risk to fuel generally bounds the lithium risk to PWSCC; hence, no “red”region is seen in Figure 3-2(b).

• “Yellow” is meant to denote control schemes where the full extent of downside riskhas yet to be determined but where recent plant-specific experience suggestssuccessful operation can be obtained using appropriate caution. Operation in the“yellow” region should occur only after a fuel and materials review has beencompleted, with appropriate fuel vendor consultation.

• “Green” is meant to denote control schemes for which the historical body ofcollective industry experience has shown successful operation to be obtainedrepeatedly. In “green” regions, observed downside risk to fuel and to ex-corematerial integrity is considered to be bounded at acceptable levels.


3-3

Increasing Lithium and/or pH



Allo

y 60

0 PW

SCC

Zirc

aloy

Cor

rosi

onR

adia

tion

Fiel

ds

Operating Regime

Operating Regime

Operating Regime

DepositedCrud Effect

Lithium Effect

}}

}

Figure 3-1Schematic Representation of the PWR Primary Chemistry Optimization Problem


3-4

Table 3-1Generic Principles for Optimization of Primary System pH(1)

1. Operate at or above pHTave = 6.9 to minimize crud deposition on fuel and crud-enhanced Zircaloy oxidation.(2,3) If plant-specific considerations requireoperation below pHTave = 6.9, a fuel surveillance program should be considered.In some cases, fuel vendor concurrence may be required to comply with thisprinciple, e.g., if operating above 3.5 ppm lithium is necessary.

2. In order to decrease corrosion product deposition on the core, plants withrecent AOA experience or with core designs susceptible to AOA shouldconsider maintaining cycle pHTave constant and as high as possible consistentwith fuel vendor limits on maximum lithium concentration and ppm-lithiumdays.

3. For initial extended operation above 2.2 ppm lithium(4) or the highest reviewedexperience to achieve a pHTave 6.9, a plant-specific fuel and materials reviewshould be performed, and a fuel surveillance program should be considered.(5)

4. Once lithium has been established consistent with Principles #1 - #3,(3) lithiumeither can be controlled so as to continue operation with pHTave = constant or itcan be maintained constant, at or below 3.5 ppm, until a specified pHTave

between 6.9 and 7.4 has been reached. Note that plant-specific constraints maylimit the maximum pH to less than 7.4. The specified pH should be selected onthe basis of plant-specific impacts on fuel and materials integrity, radiationfield control, and in the case of AOA plants, core performance. Radiationfields are expected to be lower at the upper end of this pH band. pHTave = 7.4should be considered the upper operating band.

5. Once lithium has been reduced to 2.2 0.15 ppm, then consideration may begiven to maintaining lithium constant until a specified pH which is consistentwith fuel vendor integrated lithium exposure constraints has been reachedbefore continuing with Principle #6 below. This specified pHT (between 6.9and 7.4) should be selected on the basis of plant-specific impacts on fuel andmaterials integrity and on radiation field control.

6. For [Li] > 3.0 ppm, maintain lithium at the target value within 5% of lithiumconcentration until lithium reaches 3.0 ppm, and then at 0.15 ppm until [Li]decreases to 1.25 ppm. For [Li]<1.25 ppm until the end of the operating cycle,maintain lithium at the target value within 12% of lithium concentration.Lithium variations have a greater effect on pH at lower boron concentrations(see Figure 3.3).


3-5

7. Attempt to minimize pH fluctuations during power operation. During powerlevel changes, some fluctuations in pH may be unavoidable due to thesensitivity of pHTave on temperature. As explained in Appendix F,implementing the selected lithium/boron program using pH calculated at afixed reference temperature (e.g., pH300°C) instead of at Tave would eliminate thelithium addition and removal operations associated solely with fluctuationsin the dissociation constant of water when evaluated at Tave.

______________

1. Tave equals the average of Tinlet and Toutlet.

2. Corrosion rates increase by a factor of 2 to 3 at pHT 6.5 compared to the minimumrecommended pHT of 6.9. A pHT > 6.9 should be established as soon as possibleafter power ascension has begun.

3. To ensure that industry and research data are comparable, a standard approach isrecommended for calculation of pHT. EPRI has produced a family of computercodes called EPRI chemWORKS™. These codes contain a Primary System pHCalculator and data set. The pH Calculator has been shown to be consistent withthe equation set of Appendix A over the range of lithium and temperature seen inpractice.

4. Maximum with control band is 2.2 + 0.15 ppm lithium.

5. Operation with lithium up to 3.5 ppm associated with 18 to 21 month fuel cycles iswithin the bounds of utility experience. Operating experience with lithium > 3.5ppm is limited as of this writing to brief periods where xenon equilibrium wasbeing established at the beginning of core life. Therefore, longer time periods with[Li] > 3.5 ppm should be considered developmental since there is no consensus ondownside risk to fuel and materials integrity. This principle also applies tooperation with greater lithium exposure for the purpose of further extending fuelcycles (e.g., to 24 months).


3-6

*This figure shows pH at 300°C. See Text for explanations of “red”, “yellow” and “green.”

Figure 3-2aSchematic of Fuel Integrity Risk for General RCS pH Regimes*


3-7

*This figure shows pH at 300°C. See Text for explanations of “red”, “yellow” and “green.”

Figure 3-2bSchematic of PWSCC Risk for General RCS pH Regimes*


3-8

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

01002003004005006007008009001000

11001200

13001400

15001600

17001800

19002000

Boron Concentration, ppm

Lith

ium

Con

cent

ratio

n, p

pm

Li 2.20 ppm

pH = 6.9

pH = 7.0 pH = 7.1 pH = 7.2 pH = 7.3 pH = 7.4

Li 2.20 ppm

Li 3.50 ppm

Figure 3-2cCoordinated Chemistry (pHT = 6.9) and Modified Programs (plotted at Tave=300°C)


3-9

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

01002003004005006007008009001000

11001200

13001400

15001600

17001800

19002000

Boron Concentration, ppm

Lith

ium

con

cent

ratio

n, p

pm

Li 2.20 ppm

pH = 6.9

pH = 7.0 pH = 7.1 pH = 7.2 pH = 7.3 pH = 7.4

Li 2.20 ppm

Li 3.50 ppm

Figure 3-2dCoordinated Chemistry at Elevated pHT (>6.9)Example Li/B Programs (plotted at Tave=300°C)


3-10

Figures 3-2(c) and 3-2(d) present examples of pHTave regimes (based on a hypotheticalTave=300°C) that comply with the "Principles" listed above. Plant personnel shouldselect the control approach that is most consistent with their goals. Figure 3-2(c)illustrates some recommended approaches in prior revisions of these Guidelines whichremain consistent with the Principles adopted in this revision:

(i) Adopt a minimum pHTave = 6.9, a maximum lithium concentration of 2.2 ppm foroperation below 1200 ppm boron, and a maximum pHTave = 7.4.

(ii) Adopt a minimum pHTave = 6.9 and a maximum lithium of 3.5 ppm. MaintainpHTave = 6.9 until the lithium concentration reaches 2.2 + 0.15 ppm, implyingacceptable operation at elevated lithium concentrations within the pHTave = 6.9band. Then raise pHTave at constant lithium until 7.4 is reached. At the end ofthe cycle, a maximum pHTave of 7.4 will maximize radiation control benefitsbased upon soluble activity transport arguments given in Section 2.5.

(iii) Operation at pHTave = 6.9 over the whole boron range may lead to higher fields,but may be considered for plants with significant incidence of PWSCC andwhere chemistry changes are not desired.

(iv) Operation below the pHTave = 6.9 band is not recommended since increasedcorrosion product deposition on the fuel, with the risk of increased claddingtemperatures, concentration of lithium in the oxide, and enhanced Zircaloyoxidation, is expected.

Figure 3-2(d) illustrates lithium/boron programs consistent with operating beginningof life at the highest pH possible within fuel vendor constraints (and no greater than 3.5ppm lithium, except for short periods with fuel vendor concurrence) and thenoperating at constant pHT throughout the cycle. This approach is thought to hinder theprocess of corrosion product deposition onto the core in plants experiencing axial offsetanomaly and is consistent with the principles previously outlined. Although some dataindicating reduced radiocobalt levels at three stations have been presented (105), amore complete experience base for this constant, elevated pH is still being developed.


3-11

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

Delta

pH(

300C

) with

in th

e fu

ll ba

ndab

ove

pH(3

00) =

6.9

2000 1800 1600 1400 1200 1000 800 600 400 200 0Boron Concentration, ppm

Figure 3-3Variation of pH with Boron Concentration when Lithium is Maintained within +10%(2000-1650 ppm Boron), +0.30 ppm (1650-719 ppm Boron) and +24% (719-0 ppmBoron) Control Band for pH300°C = 6.9

3.3 Control and Diagnostic Parameters

Primary chemistry guideline parameters are divided into two categories as definedbelow:

• Control Parameters are those parameters which require strict control due to materialintegrity considerations. Some of these parameters are also addressed in individualplant Technical Specifications.

• Diagnostic Parameters are those parameters which assist the chemistry staff ininterpreting primary coolant chemistry variations, or which may affect radiationfield buildup, corrosion performance of system materials, or fuel integrity.


3-12

3.4 Guideline Recommendations

3.4.1 Goals

The following goals were used to establish chemistry control parameters and to developrecommendations relative to the need to take corrective action if a parameter exceeds aspecified value or deviates from a specified range:

• The reason for the variation should be evaluated.

• Impurity concentration in the RCS should be kept to a practical and achievableminimum;

• All action levels and recommended corrective actions should be consistent withplant Technical Specifications;

• Action levels should be based on quantitative information about the effects of waterchemistry variables on the corrosion behavior of RCS materials, fuel claddingcorrosion, and radiation buildup. In the absence of quantitative data, prudent andachievable action level values should be recommended; and

• Control and diagnostic parameters should be measurable at the recommendedlevels by methods that provide reliable results.

These Guidelines recommend chemistry control parameters and action level values.Where published data cannot be presented to justify treating a particular parameter asa control parameter, it is included in the Guidelines as a diagnostic parameter. Alsocontained in this section is a suggested minimum frequency of analysis, although thefrequency of measurement of some Diagnostic Parameters can be plant-specific.

3.4.2 Corrective Actions

When a parameter or value varies from normal, the reason for the variation should beevaluated. When a parameter exceeds an action level value, corrective actions shouldbe implemented. These corrective actions may be specific to each parameter and plant.Each plant should have a predefined course of action which has been developed withattention to specific concerns. The following actions are typical:

• Verify that the reactor coolant purification system is in service with maximumachievable flow and that ion exchanger bed removal efficiencies are adequate.

• Assess conditions by comparing results of various analyses and results of previousanalyses for consistency and trends.


3-13

• Identify and isolate sources of impurities.

• Increase sampling and analysis frequency for short-term trending and confirmatoryanalyses of critical chemistry parameters.

3.5 Definitions of Terms Used in the Guidelines

3.5.1 Plant Status

Three plant status situations covered in the Guidelines are defined in Table 3-2. Theyare defined relative to the thermal and hydraulic conditions within the reactor coolantsystem as they affect the chemical environment. Additionally, they have been chosento parallel the modes of operation defined in the Standard Technical Specifications.

Table 3-2Operational Status Modes

Plant Status Reactor Condition Approximate Technical SpecificationCorresponding Mode

Cold Shutdown <250°F 5 & 6Startup >250°F, but not critical 3 & 4Power Operation Reactor critical 1 & 2

3.5.2 Action Levels

The chemistry limits and action levels presented herein are appropriate for protectingsystem materials, ensuring fuel performance, and controlling radiation field buildup.Three action levels have been defined for remedial actions to be taken when parametersare confirmed to be outside the control values. Corrective actions in response toexceeding an Action Level must in all cases be consistent with plant TechnicalSpecifications.

3.5.2.1 Action Level 1

The Action Level 1 value of a parameter represents the range, outside of which data orengineering judgment indicates that long-term system reliability may be affected,thereby warranting an improvement of operating practices, or indicates increasingimpurity concentrations which warrant investigation prior to further degradation ofwater quality.

Actions to be taken if a parameter exceeds the Action Level 1 value:


3-14

• Efforts should be made to bring the parameter within the appropriate limit withinseven days.

• If the parameter has not been restored to the appropriate range within seven days, atechnical review* should be performed and a program for implementing correctivemeasures instituted. Such a program may require equipment additions ormodifications over the long term.

*It is recommended that each plant perform a formal technical review for prolonged abnormal waterchemistry conditions. The review should address an evaluation of the condition, informing appropriatelevels of management of the existence of the condition and its implications, and the possible correctivemeasures over the short and long terms.


The Action Level 2 value of a parameter represents the value, outside of which data orengineering judgment indicates significant damage could be done to the system in theshort term, thereby warranting a prompt correction of the abnormal condition.


• Efforts should be made to bring the parameter within the appropriate Action Level2 value within 24 hours.

• If the parameter has not been restored to within the Action Level 2 limit conditionwithin 24 hours, an orderly unit shutdown should be initiated and the plant shouldbe brought to a cold shutdown condition as quickly as permitted by other plantconstraints. If chemistry is improved to within the requirements of Action Level 2prior to plant shutdown, full power operation may be resumed.

• Following a unit shutdown caused by exceeding the time limit on an Action Level 2value, a technical review* of the incident should be performed and appropriatecorrective measures taken before the unit is restarted.

*It is recommended that each plant perform a formal technical review for prolonged abnormal waterchemistry conditions. The review should address an evaluation of the condition, informing appropriatelevels of management of the existence of the condition and its implications, and the possible correctivemeasures over the short and long terms.


The Action Level 3 value of a parameter represents the limit beyond which data orengineering judgment indicates that it is inadvisable to continue to operate the plant.



3-15

• An orderly unit shutdown should be initiated immediately, with reduction ofcoolant temperature to <250°F as rapidly as other plant constraints permit.*

• Following a unit shutdown caused by entering an Action Level 3 condition, atechnical review** of the incident should be performed and appropriate correctivemeasures taken before the unit is restarted.

*If chemistry is improved to within the requirements of Action Level 3 prior to plant shutdown, poweroperation may be resumed, subject to the requirements of other Action Levels and plant TechnicalSpecifications.

**It is recommended that each plant perform a formal technical review for prolonged abnormal waterchemistry conditions. The review should address an evaluation of the condition, informing appropriatelevels of management of the existence of the condition and its implications, and the possible correctivemeasures over the short and long terms.

3.6 Guideline Values for Chemistry Parameters During Power Operation

3.6.1 Reactor Coolant System Guidelines

Action level values and recommended sampling frequencies for power operation arepresented in Table 3-3. This table should be considered directive as defined inNEI 97-06. In addition, suggestions are made for preventive and corrective measuresthat should be considered for remedial actions. Frequencies and limits for parameteranalyses have been defined, but may be superseded by the requirements of plantTechnical Specifications.


3-16

Table 3-3Reactor Coolant System Power Operation Control Parameters(9) (Reactor Critical)

ControlParameter

SampleFrequency

Action Level1 2 3

Chloride, ppb 3/wk(1) (2) >150 >1,500

Fluoride, ppb 3/wk(1) (2) >150 >1,500

Sulfate, ppb 1/wk(3) (2) >150 >1,500

Lithium, ppm 3/wk(4) (5) --- ---

Hydrogen, cc (STP)/kg H2O 3/wk(6) <25(7)

>50<15(8) <5

Dissolved Oxygen, ppb 3/wk(1) >5 --- >100

__________________

1. These frequencies are a minimum based on typical plant Technical Specifications. Plant-specific frequencies may vary.

2. Plant-specific administrative limits should be established.

3. Frequency should be increased if evidence of resin ingress is noted (e.g., increasing sulfateconcentrations or high filter/strainer ∆P).

4. Increased frequency of sampling is recommended during operations that may significantlyimpact lithium concentration (e.g., feed and bleed).

5. Limits should be based on plant-specific pH control program complying with the Table 3-1Principles.

6. Increased frequency of sampling is recommended during operations that may significantlyimpact hydrogen concentration (e.g., feed and bleed, purging of pressurizer vapor space,etc.) or known periods of hydrogen instability.

7. Hydrogen levels may be reduced to 15 cc/kg 24 hours before shutdown without institutingAction Level 1.

8. Corrective action is recommended if this value is not attained but no plant shutdown issuggested.

9. The control parameters in this table must be monitored and kept within the limits asindicated in order to be compliant with NEI 97-06. Plants using less conservativemonitoring frequencies and limits must document justifications for those disparities in theirStrategic Water Chemistry Plan. Plants must also follow the Principles for optimizingprimary water chemistry in Table 3-1.


3-17

Table 3-4Reactor Coolant System Power Operation Diagnostic Parameters (Reactor Critical)

Diagnostic Parameter Sample Frequency Reason for Analysis

Conductivity, µS/cm @ 25°C 1/day(1) Assess consistency with additives

pH @ 25°C 1/day(1) Assess consistency with additives

Boron 1/day(2) As required for reactivity control

Suspended Solids, ppb 1/wk To establish and trend changes frombaseline

Silica, ppb 1/wk Plant-specific target <3,000 (3)

__________________

1. This parameter is recommended daily since it may identify the presence of othercontaminants detectable only by more time-consuming analyses.

2. Based on station-specific Technical Specifications or needs of plant operators. Plants within-line boronometers may require less frequent laboratory analyses.

3. Subject to concurrence by fuel vendor.


3-18

Corrective Action Guidelinesfor Tables 3-3 and 3-4

Parameter Out of Range Corrective ActionsChloride/Fluoride 1. Check ion exchange beds in purification system for

flow and removal efficiencies.2. Check for high RCS ammonia or conductivity

that may cause chloride or fluoride release fromresin.

3. Check makeup water purity.4. Isolate makeup water source or change to

alternate source if required.5. Identify other potential sources of fluoride.

Suspended Solids 1. Monitor RCP seal filters and reactor coolant filters for increasing ∆P.

2. Ensure adequate RCS dissolved hydrogen.Sulfate 1. Check for indications of resin released from

the purification system.2. Check ion exchanger removal efficiency and

isolate if necessary.3. Check makeup water purity

Lithium 1. Check to ensure reactor coolant dilution or boration is not in progress.

2. Ensure that a lithiated ion exchange bed is inservice.

3. Verify flow through cation or other purificationion exchangers.

4. Check for high RCS ammonia that may causelithium release from resin.

5. Adjust lithium as required to bring withinstation control program.

Hydrogen 1. Ensure appropriate hydrogen overpressure in the VCT.

2. Ensure adequate letdown flow.3. Check for indications of leaking valves or check

for air ingress to the charging system.4. Assess pressurizer steam leakage.

Dissolved Oxygen 1. Verify hydrogen concentration.2. Check for air leaks into CVCS.


3-19

3.6.2 Special Considerations

3.6.2.1 Axial Offset Anomaly

Over the past few years, several PWR operators have experienced axial offset anomalies(AOA) associated with high boiling duty core designs. The AOA problem, if notcontrolled, may require power reductions and/or plant shutdown. The detailedmechanism of AOA in operating plants is not completely understood. However, itappears that AOA is associated with the local concentration of boron in crud depositson fuel rods with significant subcooled boiling (106). Crud deposits will enhance thelocal concentration factor on the fuel and can lead to increased fuel surface claddingtemperatures (see Section 2.4). Primary system chemistry changes designed tominimize the deposition of crud on fuel surfaces are expected to reduce the severity ofthe problem, but may not be sufficient to correct the problem.

As noted in Section 3.2.1, operation above a pHTave of 6.9 to a maximum of 7.4 (with therequirement that lithium not exceed 3.5 ppm except for short periods with fuel vendorconcurrence) is recommended based on corrosion product solubility considerations andex-core radiation fields. Minimizing corrosion rates of system materials is consideredespecially important for plants susceptible to AOA. Corrosion product release rates arereduced as the pH (at temperature) is increased from 6.9 to 7.4. The change in releaserate becomes less dependent on pH as the pH approaches 7.4. This dependence isobserved for both stainless steel and Alloy 600, although the effect is more pronouncedfor Alloy 600 (107).

Alloy 600 or Alloy 690 tubing in the steam generators forms the largest surface area incontact with primary water within the pressure boundary. At cold leg temperatures,the increasing acidity of the primary water locally will produce regions that experiencehigher release rates than at Tave. Release of solvated nickel and iron-based ions into thewater from the generators is the main source of corrosion products which precipitateonto the core. Additionally, the release rates are expected to be highest at thebeginning of the cycle and decrease as protective oxides on system materials stabilize.

These considerations indicate that corrosion product generation and transport is likelyto be greatest during the earliest parts of the fuel cycle when the pH is lowest undercurrent practices. For this reason, within fuel and materials integrity constraintsdiscussed in Section 2, plants with elevated concern for the rate at which fuel depositsform on the core, especially those with core designs exhibiting significant subcoolednucleate boiling, should consider a lithium-boron program which elevates pHTave earlyin the cycle. Current fuel corrosion models suggest that the fuel cladding corrosion rateis proportional to the integrated lithium exposure (e.g. ppm x days) during a fuel cycle.The effect of lithium and pH on PWSCC of Alloy 600 is believed to be of second order.Therefore, it is anticipated that any change to the pH control program which does not


3-20

increase the total cycle exposure to lithium would be within the guidance provided inthis document (Section 2) and would not require additional fuel and materials review.

For the reasons given above, PWR operators should consider maintaining a constantpH throughout the cycle, as high as possible up to a pHTave of 7.4. If this results in anincrease in the cycle lithium ppm x days, concurrence from the fuel vendor should beobtained. Plants with high susceptibility to PWSCC also may want to consider theimpact on PWSCC susceptibility of higher lithium and pH. Some utilities have alreadybegun to move towards a constant pHTave of 7.1-7.2.

It is believed that these proposed pH programs will lead to reduced crud deposition onthe fuel. However, the effect of long term elevated pH operation, extended cycles andaggressive fuel strategies on shutdown dose rates will have to be closely monitored.

3.6.2.2 Extended Fuel Cycles

It is important to emphasize that the current movement to longer fuel cycles hasincreased the importance of the utility performing a thorough pH optimization perSection 4. Typically, 12 month fuel cycles begin with no more than 1200 ppm boron atthe start of a cycle, so a maximum of 2.2 ppm lithium is required to satisfy the pHT = 6.9requirement. Longer fuel cycles of 18 or 21 months (with the accompanying elevatedboron concentrations for significant portions of the cycle) require higher lithium levelsto maintain pHTave at least 6.9. The optimization principles recommend operation withpHTave > 6.9 despite the fact that such operation may increase fuel cladding oxidation ata fixed cladding temperature. As mentioned in Section 2, this choice is based on thefact that the possible adverse lithium effects (i.e., the effect of operation at >2.2 ppm Li,but less than 3.5 ppm) is deemed to be less important than the fuel crudding concern(i.e., the effect of operation at pHTave < 6.9), since lithium can concentrate in thickcorrosion product deposits. Furthermore, for plants concerned about early cycledeposition on the fuel, operation below pHT = 6.9 is not recommended.

3.6.2.3 Operation With Greater Than 2.2 ppm Lithium

Before operation above 2.2 ppm lithium for extended periods (e.g., >3 months), utilitypersonnel should conduct a plant-specific fuel and materials review (e.g., covering fuelcladding corrosion and PWSCC). This review could be conducted as part of the plant-specific optimization process discussed in Section 4.

The materials review should consider the following factors:

• Prior experience with PWSCC of Alloy 600 components.

• Remedial actions implemented to control PWSCC of Alloy 600 components.


3-21

• Properties of Alloy 600, e.g., yield strength, tensile strength, grain size, carboncontent, heat treatment, and microstructure.

• Fabrication procedures (e.g., mechanical expansion, explosive expansion, hydraulicexpansion, bending, grinding, and welding) and their effect on residual stresses.

• Analysis of operating stresses and their interaction with residual stresses.

• Operating temperature of Alloy 600 components.

These factors should be taken into account to evaluate the risk of increased PWSCCresulting from operation between 2.2 and 3.5 ppm lithium and to weigh this riskagainst the benefits to be derived from such operation.

In addition to the materials corrosion-susceptibility review, a review of industryexperience should be performed to determine if the intended fuel conditions andproposed lithium program might lead to a fuel corrosion concern. Items to beconsidered in such a review include:

• Fuel cladding temperatures (including the existing fuel crud loading)

• Extent of subcooled boiling

• Cladding oxide observations

• Previous levels of fuel crudding as indicated by fuel surveillance data or operationalaxial offset information.

• Fuel design and cladding composition

• Burnup

• Core management philosophy

A fuel surveillance program may be necessary if industry experience does not envelopethe fuel operating conditions. A typical fuel surveillance program consists of oxidethickness measurements on fuel rods (of different burnups), both before and afterelevated lithium operation. As discussed in Section 2.4, the current industry experiencebase includes operation up to 3.5 ppm lithium.

3.6.2.4 Chemistry Program Modification

There are limited quantitative data on the short-term impact of changing chemistryprograms but the following items should be considered:


3-22

• It is preferable to introduce a change in lithium regime after a refueling outage.Some crud transport studies indicate that the short-term effect of raising pHT may beto transport activity from the core to the steam generator. Over a full cycle, part ofthe effect of increased transport to out-of-core surfaces will be offset by decay.

• Changing the pH control regime during mid-cycle may induce crud bursts ifcrudded fuel is present in the core. In such circumstances, changes in pH controlregimes should be delayed until the crudded fuel is removed.

3.6.3 Primary System Makeup Water

The main purpose of the primary water storage system is to act as a source ofdemineralized water for boric acid solution preparation necessary for reactor coolantmakeup. The system also provides makeup water to other systems and serves as asource for flushing and decontamination of components.

Selected diagnostic parameters are summarized in Table 3-5. Analysis frequencies forsilica, calcium, magnesium and aluminum should be consistent with fuel vendorrecommendations.

It is also important to note that oxygen depletes hydrogen inventory. Minimizingmakeup water dissolved oxygen will minimize any potential effects that may occurduring heavy periods of dilutions. An additional consideration is that, with an aeratedsystem, nitrogen is added with oxygen leading to the formation of ammonia in theRCS. Large additions could cause some lithium throw from the CVCS beds and affectthe pH.

The following table does not attempt to present all makeup water parameters.

Table 3-5Source Water for Reactor Makeup(1) Diagnostic Parameters -- All Modes

Diagnostic Parameter Frequency Expected Value(2)

Reactive Silica, ppb 1/qtr <100Aluminum, ppb 1/qtr <80Magnesium, ppb 1/qtr <40Calcium, ppb 1/qtr <40Oxygen, ppm 1/qtr Variable

_______________1. This table is not intended to include all makeup water parameters, but is focused on

those hardness elements which impact fuel performance.2. Typical plant values are much lower than the "less than" values shown in the table.


3-23

3.7 Control and Diagnostic Parameters, Frequencies, and Limits for StartupChemistry

The monitoring parameters, frequencies and limits for RCS chemistry during startupare given in Tables 3-6, 3-7, and 3-8. Table 3-6 and 3-7 should be considered directivesas defined in NEI 97-06. Startup and shutdown chemistry practices and theiroptimization are discussed in Volume 2 of these Guidelines.

Table 3-6Reactor Coolant System Cold Shutdown Control Parameters (Reactor <250°F)

Control Parameter Sample Frequency(1) Value Prior to Exceeding 250°F

Chloride, ppb 1/wk <150

Sulfate, ppb 1/wk <150

Fluoride, ppb 1/wk <150

Oxygen, ppb (2) <100

__________________

1. These frequencies are minimum. Technical Specifications should be consulted.Frequencies should be increased during activities such as refueling when the likelihood ofcontamination is increased.

2. Applicable only during heatup. C.E. and B&W plants may have a requirement to have <100 ppb prior to 150oF or a N2H4 residual. Pressurizer oxygen should also be < 100 ppbprior to exceeding 250°F pressurizer temperature. Pressurizer sampling can be terminatedwhen dissolved oxygen is < 100 ppb. Some plants may not be able to sample thepressurizer liquid space, and must rely on volume turnover or hydrazine addition toensure that the dissolved oxygen is in specification. In any case, individual plantrequirements based on Technical Specifications or NSSS requirements apply.


3-24

Table 3-7Reactor Coolant System Startup Control Parameters(1) (Reactor Subcritical and >250°F)

Control Parameter Sample Frequency Value Prior to Criticality

Chloride, ppb (1) <150

Sulfate, ppb (1) <150

Fluoride, ppb (1) <150

Hydrogen,cc (STP)/kg H2O

(1) >15(2)

Dissolved Oxygen, ppb (1) <100

__________________

1. Sample frequency shall be established by chemistry management based on plant startupschedule and Technical Specification requirements.

2. Hydrogen overpressure to be established prior to power ascension. Plant-specificconsiderations should be evaluated to minimize the likelihood of delays because ofdegassing requirements in the event of an unexpected shutdown.


3-25

Table 3-8Reactor Coolant System Startup Chemistry Diagnostic Parameters (Following Fill-and-Vent to Reactor Critical)

Parameter Frequency(1) Note(s)

Boron 1/day(1) As required for reactivity controlpH25°C 1/dayConductivity µS/cm @25°C 1/dayLithium 1/day(2,3) Consistent with Station ProgramCorrosion products(Fe, Ni, 58Co, 60Co, 54Mn, 51Cr)

1/day(3) (4)

Suspended Solids, ppb 1/day(3) Should be <350Ammonia, ppb 1/72 hrs.(3)

RCS temperature 1/day(3)

Letdown flow 1/8 hrs.Purification demin. DF 1/day(5)

Silica, ppb 1/day(3)

Hydrazine as required

__________________

1. Consistent with station Technical Specifications.

2. Sampling frequency should be increased as necessary to establish or re-establish theparameter within the station program parameters.

3. Sampling frequency should be increased (e.g., hourly or 1/4 hrs.) as startupconditions warrant to profile cleanup efforts. Suspended solids should bedetermined before criticality.

4. Analyze elemental Fe and Ni from >150°F until 500°F. Samples may be drawn on apredetermined schedule and preserved for analysis at a later time. However, plant-specific programs for managing corrosion product behavior during heatup (Section4.2 of Volume 2) may require immediate analyses for nickel.

5. Demineralizer surveillance should be performed to ensure continued removal ofimpurities during heatup. Surveillance of CVCS demineralizers is especiallyimportant following chemical additions to ensure the absence of chloride or sulfatethrow from resins.

4-1

4 METHODOLOGY FOR PLANT-SPECIFIC

OPTIMIZATION

4.1 Introduction

The purpose of this section is to provide a framework for plant chemistry personnel todevelop an optimized primary chemistry program considering plant design, materialsof construction, fuel design and cycle length, corrosion degradation history, regulatorycommitments, fuel vendor warranty requirements, etc. This section outlines thedocumentation which each utility must develop to define a Strategic Water ChemistryPlan. Documentation is a requirement of these Guidelines per the NEI SG Initiative 97-06.

The recommended approach recognizes that nuclear power plants must consider avariety of issues in developing a primary water chemistry program and that theseissues must be dealt with on a plant specific basis. The approach is similar to theoptimization methodology presented in the EPRI-published PWR Secondary WaterChemistry Guidelines – Revision 4, TR-102134, Rev. 4, Nov. 1996. Optimization of theprimary water chemistry program is the process of developing a program whichappropriately reflects the technical bases provided in Section 2 relative to approachesfor insuring primary system and fuel integrity and minimizing radiation fields.

In addition to providing the technical bases for the primary water chemistry program,the Strategic Water Chemistry Plan will:

• Provide continuity of programmatic decisions to help personnel understand whycertain parameters or limits have been adopted, and

• Provide for periodic reviews (e.g., end-of-cycle) of technical bases to determinerequired program changes.

As discussed in Section 1, NEI 97-06 commits the industry to adopting the requirementsof these Guidelines to ensure steam generator integrity. However, not all the primarywater chemistry parameters discussed in these Guidelines are believed to impact steamgenerator integrity. For those parameters that have an effect on steam generator

Methodology for Plant-Specific Optimization

4-2

integrity, the optimization process includes adopting a site-specific limit within thebounds established in Tables 3-3, 3-6 and 3-7. For example, chemistry personnel couldadopt the a 50 ppb chloride limit as the station Action Level 1 value if a site-specificevaluation suggests that operation above this limit reflects ion exchanger deterioration.Chemistry personnel must document such decisions and include a brief description ofthe bases for the decision.

For those parameters that do not affect steam generator integrity, chemistry personnelhave additional optimization flexibility. For example, Table 3-4 and Section 2 shouldbe considered when optimizing each RCS parameter during power operation withoutspecifying an acceptable range. For example, RCS silica concentration limits must“balance” the possible effects on fuel deposits and the costs of cleanup. The 1 ppmvalue discussed in previous guidelines had been presented because it was included inmany fuel warranty documents. This value can be adjusted in the optimization processbased on the utility evaluation of risks of elevated silica operation and the costs of pre-startup cleanup. Such an optimization philosophy can be implemented for allparameters that do not have a direct materials integrity impact.

The results of the optimization process should be incorporated into a station chemistryprogram that is approved for implementation by plant management. The completionof a “Strategic Water Chemistry Plan” is a requirement of these Guidelines (and NEI 97-06).

It should be noted that the “Strategic Water Chemistry Plan” is a living document thatshould be intermittently reviewed against plant experience and research data todetermine if modifications should be made.

4.2 Primary Water Chemistry Variables and Options

4.2.1 Program Objectives

As noted in Section 2, the purpose of the primary water chemistry control program isto:

• Ensure primary system pressure boundary integrity,

• Ensure fuel cladding integrity and achievement of design fuel performance, and

• Minimize out-of-core radiation fields.

The optimum chemistry control approach for accomplishing one of these goals may notbe the optimum approach for another. Therefore, chemistry programs may reflect acompromise based on station-specific priorities. The highest priorities will always be


4-3

the water chemistry approaches that ensure pressure boundary integrity and fuelcladding integrity. In Section 3, the parameters that affect structural integrity are listedas Control Parameters and are presented with Action Level values and correctiveactions. A lower priority for the primary water chemistry program is radiation fieldcontrol. The parameters affecting radiation fields, noted in Section 3 as DiagnosticParameters or Control Parameters without Action Level values, have a much widerrange for site-specific optimization.

4.2.2 Parameters Impacting Structural Integrity

Operational ranges and Action Level limits for parameters deemed to have anestablished impact on primary system integrity have been specified in Tables 3-3, 3-6and 3-7. Chemistry personnel may adopt specific values within these predefinedranges without technical justification because the ranges identified are consistent withthe technical bases noted in Section 2. Chemistry personnel may also adopt valuesoutside the ranges presented in Section 3 if justified by a formal evaluation or ifconsistent with plant Technical Specifications. This evaluation must be documentedand approved by station management in the “Strategic Water Chemistry Plan.”Utilities may also adopt Action Level responses which are different than those shownin these Guidelines. The basis for each deviation must be documented.

4.2.2.1 Chloride/Fluoride/Sulfate

Many station Technical Specifications include specific limits and action levels forchloride and fluoride. These species are implicated in stress corrosion cracking ofaustenitic stainless steels and corrosion of Zircaloy, although the impacts at normalPWR operating conditions are minimal at typical chloride and fluoride concentrations.

The plant-specific chloride and fluoride operating limits and associated action levelresponses must be consistent with regulatory restrictions. The Action Levels andresponses described in Section 3 are consistent with the wording of most plantTechnical Specifications. Revised wording may be desirable for plants where TechnicalSpecifications do not address primary chemistry parameters.

Sulfate is not covered in most plant Technical Specifications. However, RCS sulfate,believed to be principally caused by resin or resin decomposition product intrusionsfrom the CVCS, has been implicated in both stress corrosion cracking and intergranularattack of Alloy 600 CRDM and other vessel head penetrations. This evaluation issummarized in NRC Generic Letter 97-01. Current sulfate control limits have been setat concentrations consistent with chloride and fluoride limits, and are considered toprovide adequate protection against stress corrosion cracking and intergranular attack.


4-4

Action Level values or hold limits are provided in Tables 3-3, 3-6 and 3-7 for chloride,fluoride and sulfate. Plant-specific limits lower than these values can be adoptedwithout a detailed evaluation. Limits greater than these guideline recommendationsmust be based on a formal plant-specific technical evaluation.

4.2.2.2 Dissolved Oxygen Control

Dissolved oxygen is controlled in the RCS to avoid oxidizing conditions and corrosionproduct generation associated with loss of hydrogen inventory. For structural integrityconcerns, oxygen in the RCS must be controlled within the limits of Tables 3-3, 3-6 and3-7. Oxygen can also be controlled in auxiliary systems to minimize oxygenintroduction into the coolant. These systems include makeup water, boric acid storagetanks and the RWST. The decision on whether or not to include oxygen control onthese systems should be based on expected costs and impacts on corrosion andcorrosion product transport.

Action Level values or hold limits are provided in Tables 3-3, 3-6 and 3-7 for reactorcoolant dissolved oxygen. Plant-specific limits lower than these values can be adoptedwithout a detailed evaluation. Limits greater than these guideline recommendationsmust be based on a plant-specific evaluation and documented.

4.2.2.3 Hydrogen

Hydrogen is generally controlled within the limits of the Table 3-3. The traditionallimit of 25-50 cc/kg has been shown to be adequate to achieve protection againstradiolytically-produced oxidants as well as minor operational upsets. Plant-specificlimits different from those in Table 3-3 may be selected based on some of the followingconcerns:

• Steam generator tubing PWSCC

• Pressure limits on VCT

• Concentrations of hydrogen desired prior to shutdown (The concentration can affectthe time required for a plant to degas during a shutdown/cooldown.)

Action Level values or hold limits are provided in Tables 3-3 and 3-7 for reactor coolantdissolved hydrogen. Plant-specific limits within these values can be adopted without adetailed evaluation. Limits outside these guideline recommendations must be based ona plant-specific evaluation and documented.


4-5

4.2.2.4 pHT (Lithium Control Parameter)

The plant-specific pHT program can be impacted by NRC regulatory commitments,fuel vendor warranty requirements, PWSCC concerns, fuel design or cycle length, AOAsusceptibility or overall plant economic considerations (i.e., the amount of chemicalsrequired over the cycle). An optimized pHT program is designed to minimize fueldeposits, fuel cladding and system materials corrosion, and out-of-core radiation fields.

• Regulatory Commitments: The pHT approach currently in use at many PWRs isaddressed in Technical Specifications or FSARs. Chemistry personnel must ensurethat the selected approach is consistent with regulatory commitments.

• PWSCC Concerns: Section 2 discusses the impact of pHT and lithium on Alloy 600SCC. Though the impacts are believed to be minor, the selection of pHT controlregimes will depend on plant materials and corrosion history, particularly PWSCC.

• Fuel Vendor Warranty Requirements: Many fuel vendors provide lithium limitswithin their warranty. These limits may be in ppm (i.e., maximum lithiumconcentration), ppm-days (i.e., total integrated lithium exposure), etc. Vendor limitsmay also include a minimum coolant pHT. The selected pHT control approachshould be consistent with vendor limits.

• Fuel Design or Cycle Length: The pHT control approach can also be affected by fueldesign parameters (e.g., boiling void fraction, amount of burnable poisons, etc.) orcycle length (i.e., high boric acid concentrations at BOL). These issues need to beconsidered to ensure that minimum pHT requirements are met without exceedingtotal integrated lithium exposure.

• Plant Economic Issues: Costs of RCS chemicals and resin can impact on the pHT

control strategy during transients. This issue is considered much less significantthan materials/fuel corrosion issues.

4.2.2.5 Zinc

Laboratory tests suggest that zinc addition to the reactor coolant can be of benefit withrespect to PWSCC. Zinc additions also have been shown to decrease out-of-coreshutdown dose rates. However, plant operating data are limited and there arepotential negative impacts on fuel cladding oxidation rate.

4.2.3 Parameters With Negligible Impact on Structural Integrity

The following parameters are believed to have no direct impact on steam generatorintegrity. For these parameters, operational ranges may be suggested in Section 3, but


4-6

no Action Level values are presented. Chemistry personnel should assign specificoperational limits based on a site-specific optimization study. The optimization studyshould be consistent with the water chemistry program objectives. These evaluationsshould be documented in the Strategic Water Chemistry Plan.

4.2.3.1 Silica

In recent years, nuclear plants have experienced high silica concentrations in the RCSand spent fuel pools, often due to degradation of Boraflex. This has led to significantcleanup costs prior to operation to satisfy fuel vendor warranties or chemistry programrequirements. Some stations have performed technical evaluations to determine ifoperational silica concentrations can be relaxed and still satisfy the intent of the fuelvendor warranty. Limited relaxation may require increased program surveillance butis expected to save significant startup time and costs. The decision to adopt anincreased silica concentration limit should be based on a plant-specific evaluation ofpotential cost savings versus possible fuel risks.

4.2.3.2 Suspended Solids

Reactor coolant suspended solids are usually very low during normal power operation.A typical value of <10 ppb is observed. Suspended solids can increase as a result ofchemical, thermal or hydraulic changes that affect crud movement. Highconcentrations of suspended solids will rapidly redeposit on system surfaces, and asmall portion will be removed by the CVCS filter. There is little that can be done inresponse to detection of high suspended solids concentrations. Periodic analysesshould be performed to monitor trends.

4.2.4 Chemistry Control During Shutdowns and Startups

4.2.4.1 Shutdown Chemistry Program

Numerous shutdown chemistry program options exist. Variables that affect decisionson shutdown chemistry include: expected critical path schedule, temperature holdsassociated with other station tests or procedures, overall outage length, expectedmagnitude of corrosion product release, and method of oxidation. Shutdownchemistry control recommendations are presented in Volume 2 of these Guidelines andshould be considered when optimizing the program, both for planned outages and forforced midcycle outages. Variations in the program that do not impact on systemintegrity but that can significantly impact on transport of corrosion products arepossible.


4-7

4.2.4.2 Startup Chemistry Program

Startup chemistry controls principally should ensure that RCS impurities (halogens,sulfates and oxygen) meet Table 3-7 limits. Many aspects of plant operation duringstartup (e.g., boron sampling for criticality) are governed by other station procedures.An optimized startup chemistry program should be developed to be consistent withplant startup schedules while considering previous startup/operation/shutdownexperiences. Startup chemistry control is discussed in Volume 2 of these Guidelines.

4.3 Optimization Methodology

4.3.1 Optimization Process

These Guidelines, as a result of NEI Steam Generator Initiative 97-06, requiredevelopment of a “Strategic Water Chemistry Plan.” Utilities may choose to documentthe technical bases for their program using a wide variety of methodologies. Themethodology and basis must be defined in the “Strategic Water Chemistry Plan.” Themethodology generally will include the following steps:

1. Identify specific program objectives.

2. Establish priorities for program objectives.

3. Identify and assess positive and negative impacts of chemistry control approaches.

4. Identify chemistry program approaches which will be implemented, includingaction level limits and responses.

Table 4-1 presents a generic evaluation of primary chemistry control options relative topossible impacts on program objectives. Each program approach should be reviewedby plant personnel relative to achievement of their Strategic Water Chemistry Programgoals, considering all pertinent aspects of plant design.

4.3.2 NEI SG Initiative Checklist

Utilities should review the following list to be sure they have met the requirements ofthis guideline which involve SG tube integrity. Completing these requirements isconsidered mandatory to be in compliance with NEI SG Initiative 97-06.

1. A “Strategic Water Chemistry Plan” has been developed and documented. Thisplan has been approved by an appropriate level of station management. The basesfor the primary chemistry program, including the pHT control program, must bediscussed in the plan.


4-8

2. If any plant-specific Action Levels or sample frequencies for control parametersshown in Tables 3-3, 3-6 and 3-7 are outside the bounds of the limits of Tables 3-3, 3-6 and 3-7, the technical justification for the plant-specific parameters must beprovided in the plan.


4-9

Table 4-1Chemistry Control Program Approaches

Contaminant Minimization Elevated pHT

Positive Impacts Negative Impacts Positive Impacts Negative Impacts

SystemIntegrity

Reduced risk of SCC atlow concentrations ofchloride, fluoride, sulfate,and oxygen

(1) Decrease in corrosion andcorrosion product releaserates

Increased tendency for Alloy600 PWSCC

Fuel Integrity Reduce risk of cladcorrosion (fluoride)

(1) Decrease in claddingtemperature/oxidationrate due to decrease indeposit loading.Decreased tendency forAOA.

Possible increase in cladoxidation rate (lithium effect)Possible deposit peaking inupper regions of core

Cost for fuel surveillance (ifrequired)

Dose Rates Possible slight dose ratereduction if generalcorrosion rates decreasewith contaminantconcentration reduction

(1) Expected decrease due togeneral corrosion andcorrosion producttransport reduction

None

1. The only identified negative impact of a contaminant minimization program is a slight increase in operating costs


4-10

Table 4-1 (continued)

Shutdown/Startup Chemistry Control Silica Control

Positive Impacts Negative Impacts Positive Impacts Negative Impacts

SystemIntegrity

No expected impact oncorrosion withinrecommended parameterranges

No expected impact oncorrosion withinrecommended parameterranges

No expected corrosionimpact

Costs to reduce silicaconcentration can besignificant

Fuel Integrity Deposit removal can beenhanced by tailoringshutdown and startupprograms

None identified Possible reduction in fueldeposits if silicamaintained <1 ppm inpresence of hardnesselements.

No impact expected atsilica < 3 ppm in absenceof hardness elements.

Costs to reduce silicaconcentration can besignificant

Dose Rates Minimal expected impact,i.e., < 5% reduction ifshutdown programoptimized

Chemistry variationsoutside Guidelinesrecommendations(particularly increase in Liduring shutdown) couldlead to hot spotdevelopment

None Costs to reduce silicaconcentration can besignificant


4-11

Table 4-1 (continued)

Zinc Injection

Positive Impacts Negative Impacts

SystemIntegrity

Reduction in Alloy 600PWSCC appearsachievable based onlaboratory data

Cost can be high ifdepleted zinc approachselected

Fuel Integrity Impact on fuel corrosionhas not been welldefined. Field data after9 month applicationsuggest minimal effect.

Cost can be high ifdepleted zinc approachselected

Dose Rates Dose rate reductionsappear achievable.Greater benefit expectedif depleted zincemployed.

Cost can be high ifdepleted zinc approachselected.

5-1

5 REFERENCES

1. C. M. Owens, "The Primary-Side Stress Corrosion Cracking Performance of MillAnnealed Inconel - 600 Tubing in C-E Designed Steam Generators," NEA/CSNI -UNIPEDE Specialists Meeting on Steam Generators, Stockholm, Sweden, October 1-5, 1984.

2. S. M. Bruemmer, L. A. Charlot, and C. H. Henager, Jr., "Microstructure andMicrodeformation Effects on IGSCC of Alloy 600 Steam Generator Tubing,"Corrosion '87, Paper #88, San Francisco, Calif., March 9-13, 1987.

3. Optimization of Metallurgical Variables to Improve Corrosion Resistance of InconelAlloy 600, Palo Alto, Calif.: Electric Power Research Institute. July 1983. NP-3051.

4. Proceedings of the PWSCC Remedial Measures Workshop, Clearwater Beach, FL,Palo Alto, Calif.: Electric Power Research Institute. February 1989. NP-6719-M-SD.

5. NRC Bulletin No. 89-01: "Failure of Westinghouse Steam Generator TubeMechanical Plugs", May 15, 1989.

6. PWSCC of Alloy 600 Materials in PWR Primary System Penetrations, Palo Alto,Calif.: Electric Power Research Institute, July 1994. EPRI TR-103696.

7. Proceedings: 1997 EPRI Workshop on PWSCC of Alloy 600 in PWRs, Palo Alto,Calif.: Electric Power Research Institute. November 1997. TR-109138-P1 and P2.

8. Microstructure and Stress Corrosion Resistance of Alloys X-750, 718, and A-286 inLWR Environments, Palo Alto, Calif.: Electric Power Research Institute, June 1989.NP-6392M

9. B. M. Gordon, "The Effect of Chloride and Oxygen on the Stress Corrosion Crackingof Stainless Steels: Review of Literature." Materials Performance, Vol. 19, No. 4, pp.29-38, 1980.

10. Materials Handbook for Nuclear Plant Pressure Boundary Applications, Palo Alto,Calif.: Electric Power Research Institute. December 1997. TR-109668-SI.

References

5-2

11. Multivariable Analysis of Effects of Li, Hydrogen and pH on PWR Primary WaterStress Corrosion Cracking, Palo Alto, Calif.: Electric Power Research Institute. May1996. TR-105656.

12. Survey of Corrosion Product Generation Transport and Deposition in Light WaterNuclear Reactors. Palo Alto, Calif.: Electric Power Research Institute, March 1979.NP-522.

13. W. G. Burns. Suppression by Dissolved Hydrogen of Radiolysis of Water by MixedRadiation Fields. AERE-M2702, 1975.

14. G. H. Jenks. Effects of Reactor Operation on HFIR Coolant. Oak Ridge NationalLaboratory, 1965. ORNL-3848.

15. W. G. Burns and P. B. Moore. "Radiation Enhancement of Zircaloy Corrosion inBoiling Water Systems: A Study of Simulated Radiation Chemical Kinetics." WaterChemistry of Nuclear Reactor Systems, In Proceedings of the BNES InternationalConference, Bournemouth, UK, 1977.

16. C. Burn, et al., “Radiolysis Studies at Belleville (PWR 1300) Water Chemistry withLow Hydrogen Concentration,” presented at Chemistry in Water Reactors:Operating Experience and New Developments, Nice, France. April 1994.

17. H. Christensen, “Calculations of Radiolysis in PWR,” presented at Chemistry inWater Reactors: Operating Experience and New Developments, Nice, France. April1994.

18. R. S. Pathania and, A. R. McIlree. "The Effect of Hydrogen on Stress CorrosionCracking of Alloy 600 in 360°C Water." In Proceedings of the Third InternationalSymposium on Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors, Traverse City, Michigan, August 30-September 3, 1988.

19. G. Economy, R. J. Jacko, J. A. Begley, and F. W. Pement. "Influence of HydrogenPartial Pressure on the IGSCC Behavior of Alloy 600 Tubing in 360°C Water or400°C Steam." Paper no. 92, NACE Corrosion-1987, San Francisco, California, March9-13, 1987.

20. Effect of Lithium Hydroxide on Primary Water Stress Corrosion Cracking of Alloy600, Palo Alto, Calif.: Electric Power Research Institute. September 1991. NP-7396-S.

21. G. Economy, et al., "Temperature Effects on Alloy 600 PWSCC from 310 to 330°C,"Corrosion 89, 17-21. April 1989.

References

5-3

22. D. Briceno, et al., "The Influence of Chemical Factors on the Initiation of PrimarySide IGSCC in Alloy 600 Steam Generator Tubing," Proceedings of the InternationalSymposium Fontevraud III - Contribution of Materials Investigation to theResolution of Problems Encountered in Pressurized Water Reactors, September 12-16, 1994.

23. K. Norring, "Influence of LiOH and Hydrogen on Primary Side IGSCC of Alloy 600Steam Generator Tubes," Studsvik Energy, October 1989.

24. S. Smialowska, "Intergranular SCC of Alloy 600 in Primary Side PWREnvironments," Materials Science and Engineering Department, Ohio StateUniversity, December 1991.

25. N. Ogawa, et al., "PWSCC Susceptibility of Mill Annealed Alloy 600 in ReactorCoolant System Water During High pH Operation," Nuclear Power EngineeringCorporation, Tokyo, Japan. (Submitted to Nuclear Engineering and Design).

26. T. Cassagne et al., “An Update on the Influence of Hydrogen on the PWSCC ofNickel Base Alloys in High Temperature Water,” Proceedings: Eighth InternationalSymposium on Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors, August 10-14, 1997, Amelia Island, FL, p. 307, ANS.

27. D. D. Whyte and L. F. Picone, “Behavior of Austenitic Stainless Steel in PostHypothetical Loss of Coolant Environment.” WCAP-7803, 1971.

28. BWR Water Chemistry Guidelines - 1993 Revision. Palo Alto, Calif: Electric PowerResearch Institute, February 1994. TR-103515.

29. R. L. Jones, ed. EPRI Activities in Support of TMI-l Steam Generator Recovery.Memorandum Report, April 1983.

30. R. L. Long, Group Leader. GPUN Nuclear Technical Data Report, TMI-l OTSGFailure Analysis Report, July 1982.

31. R. Bandy, R. Roberge, and R. C. Newman. "Low Temperature Stress CorrosionCracking of Sensitized Inconel 600 in Tetrathionate and Thiosulfate Solutions."Corrosion, Vol. 39, 10, October 1983, pp. 391-398.

32. J. J. Pacios, “Defects in the Jose Cabrera NPP,” REVISTE SNE, December 1994. pp.56-60.

33. NRC Generic Letter 97-01, “Degradation of Control Rod Drive Mechanism Nozzleand Other Vessel Closure Head Penetrations,” April 1, 1997.

References

5-4

34. S. Fyfitch, “B&WOG Integrated Response to Generic Letter 97-01: Degradation ofControl Rod Drive Mechanism Nozzle and Other Vessel Closure HeadPenetrations,” BAW-2301, July 1997.

35. “CEOG Response to NRC Generic Letter 97-01, Degradation of Control Rod DriveMechanism Nozzle and Other Vessel Closure Head Penetrations,” CE NPSD-1085,Revision 0, July 1997.

36. Individual Westinghouse utility responses to GL 97-01, filed in NRC PublicDocument Room under the following NRC accession numbers: Beaver Valley:9708040131 970730; Braidwood: 9708040212 970730; Byron: 9708040212 970730;Callaway: 9707230079 970717; Catawba: 9708060052 970730; Comanche Peak 1 & 2:9708050243 970729; Cook 1 & 2: Diablo Canyon: 9707310234 970728; Farley 1&2:9707300135 970725; Ginna: 9707310152 970725; Harris: 9708050318 970729;9711140289 971104; Indian Point 2: 9708190335 970730; Indian Point 3: 9707290089970721; Kewaunee: 9708060216 970730; McGuire: 9708060052 970730; Millstone 3:9707310015 970728; North Anna 1 & 2: 97080100003 970725; Prairie Island:9708050182 9707289; Robinson: 9708040125 970729; Salem: 9708050258 970730;Seabrook: 9708050359 970730; Sequoyah 1 & 2: 9708050173 970730; South TexasProject: 9708050242 970729; Summer: 9708060066 970730; Surry 1 & 2: 9708010003970725; Turkey Point: 9708060017 970728; Watts Bar 1: 9708050173 970730; Vogtle 1& 2: 9707310054 970724; Zion: 9708040212 970730.

37. Crack Growth and Microstructural Characterization of Alloy 600 PWR Vessel HeadPenetration Materials, Electric Power Research Institute, Palo Alto, Calif.: December1997, TR-109136.

38. R. S. Pathania, B. Cheng, M. Dove, R.E. Gold, and C. A. Bergmann, “Evaluation ofZinc Addition to the Primary Coolant of Farley-2 Reactor,” Proceedings: EighthInternational Symposium on Environmental Degradation of Materials in NuclearPower Systems-Water Reactors, August 10-14, 1997, Amelia Island, FL, p. 379, ANS.

39. R. S. Pathania, B. Cheng, M. Dove, R. E. Gold, and C. A. Bergmann, “Evaluation ofZinc Addition to the Primary Coolant of Farley-2 PWR,” Fontevraud IVInternational Symposium, Contribution of Materials Investigation to the Resolutionof Problems Encountered in Pressurized Water Reactors, September 14-18, 1998.

40. Zircaloy Corrosion Properties Under LWR Conditions (Part 1), Electric PowerResearch Institute, Palo Alto, Calif. October 1990. NP-6979-D.

41. Y. Solomon and J. Roesmer. Nuclear Technology, Vol. 29, 1976, p. 166.

42. Y. L. Sandler. Corrosion, Vol. 35, 1979, p. 205.

References

5-5

43. R. Riess, Nuclear Technology, Vol. 29, 1976, pp. 153-159.

44. G. P. Sabol, et al., Rootcause Investigation of Axial Power Offset Anomaly, ElectricPower Research Institute, Palo Alto, Calif.: TR-108320, June 1997.

45. P. Cohen., Water Coolant Technology of Power Reactors, American Nuclear Society,1980.

46. M. Lippens, et al., "Fuel Rod Failure Induced by Waterside Corrosion in Reactor BR-3. Failure Description and Water Chemistry Analysis," IAEA Meeting on theInfluence of Power Reactor Water Chemistry on Fuel Cladding Reliability, SanMiniato, Italy, October 1981.

47. Berry, White, and Fink, Corrosion, Vol. 19, 1963, p. 253.

48. A. Strasser, et al., Corrosion Product Buildup on LWR Fuel Rods, Palo Alto, Calif.,Electric Power Research Institute, April 1985. NP-3789.

49. B. Cheng, "RCS Silica and Fuel Reliability," Presentation at the EPRI PWR PrimaryWater Chemistry Guidelines Meeting, Clearwater, FL, February 4, 1998. See also"Effect of Silica on Fuel in PWRs", EPRI, August 26, 1996.

50. J. Roesmer, "Minimizing Core Deposits and Radiation Fields in Pressurized WaterReactors by Coordinated Li/B Chemistry." Third International Conference onWater Chemistry of Nuclear Reactor Systems, Bournemouth, UK, October 1993, pp.29-36.

51. Y. Solomon, "An Overview of Water Chemistry for Pressurized Water NuclearReactors." First International Conference on Water Chemistry of Nuclear ReactorSystems, Bournemouth, UK, October 1977.

52. Seminar on PWR Primary Water Chemistry and Radiation Field Control -Proceedings. Palo Alto, Calif.: Electric Power Research Institute, September 1983.

53. Coolant Chemistry Effects on Radioactivity at Two Pressurized Water ReactorPlants. Palo Alto, Calif.: Electric Power Research Institute, March 1984. NP-3463.

54. E. Hilner and J. N. Chirigos, "The Effect of Lithium Hydroxide and RelatedSolutions on the Corrosion Rate of Zircaloy in 680° F," August 1962. WAPD-TM-307.

55. H. Coriou, et al., "The Corrosion of Zircaloy in Various Alkaline Media at HighTemperature," Corrosion of Reactor Materials II, IAEA, Vienna, 1962, p.193.

References

5-6

56. S. G. McDonald, et al., "Effect of Lithium Hydroxide on the Corrosion Behavior ofZircaloy-4," Zirconium in Nuclear Industry: Sixth International Symposium. ASTMSTP824, 1984. pp. 519-530.

57. W. K. Anderson and M. J. McGoff, "Corrosion of Zircaloy in Crevices UnderNucleate Boiling Conditions," KAPL-2203, April 13, 1962.

58. W. K. Winegardner, "Corrosion in Simulated PRTR Fuel Element Surface Crevices",BNWL - 83, July, 1965.

59. S. H. Shann, et al., "Effects of Coolant Li Concentration on PWR Cladding WatersideCorrosion," Proceeding, ANS/ENS International Topical Meeting on LWR FuelPerformance, Avignon, France. April 21-24, 1991.

60. G. Sabol, et al., "Zirlo Performance Update and Corrosion Modeling," Proceedings,ANS International Topical Meeting on Light Water Reactor Fuel Performance, WestPalm Beach, FL. April 17-21, 1994.

61. J. T. Willse, et al., "Recent Results From the B&W Fuel Company's Fuel PerformanceImprovement Program," Proceedings, ANS International Topical Meeting on LightWater Reactor Fuel Performance, West Palm Beach, FL. April 17-21, 1994.

62. B. Cheng, P. M. Gilmore and H. H. Klepfer, “PWR Fuel Cladding CorrosionPerformance Mechanisms and Modeling,” presented at ASTM EleventhInternational Symposium on Zirconium in Nuclear Industry, Garmisch, Germany.September 11-15, 1995.

63. EPRI PWR Fuel Cladding Corrosion (PFCC) Code Manual, Electric Power ResearchInstitute, Palo Alto, Calif.: October 1995. TR-105387, Vol. 1&2.

64. P. Billot, et al., "Experimental and Theoretical Studies of Parameters that InfluenceCorrosion of Zircaloy-4," Zirconium in Nuclear Industry: Tenth InternationalSymposium. ASTM STP1245, 1994. pp. 351-377.

65. R. A. Weiner, et al., "The Effects of Coolant Chemistry Control and FuelManagement Strategies on Fuel Cladding Corrosion," paper presented at TopFuel'97 Conference, Manchester, England, June 1997.

66. A Comparison of Zircaloy Oxide Thickness on Millstone-1 and North Anna-1 PWRFuel Cladding, Electric Power Research Institute, Palo Alto, Calif.: August 1993.TR-102826.

67. M. V. Polley, Evaluation of the Impact of Elevated Lithium Concentrations onZircaloy-4 Fuel Clad Oxidation in the Millstone-3 PWR. Palo Alto, Calif.: ElectricPower Research Institute, December 1995. TR-105662.

References

5-7

68. M. V. Polley, "Assessment of Zircaloy Fuel Clad Oxidation Under Elevated LithiumPrimary Coolant Chemistry at Millstone 3 PWR", Proceedings of the InternationalConference on Chemistry and Water Reactors, Nice, France, Vol. 1, 1994, p. 82.

69. E. Picard, et al., Zircaloy Corrosion in PWRs: Effect of Additives, Palo Alto, Calif.:Electric Power Research Institute, July 1996. TR-104517.

70. I. L. Bramwell, et al., "Corrosion of Zircaloy-4 PWR Fuel Cladding in Lithiated andBorated Water Environments," Zirconium in the Nuclear Industry: NinthInternational Symposium, ASTM STP1132, 1991. pp. 628-641.

71. J. Thomazet, et al., "In Reactor Corrosion and Crud Deposition Data on FragemaFuel", Proceedings, ANS Topical Meeting on Light Water Reactor Fuel Performance,Vol. 1, Orlando, FL, April 21-24, 1985, pp. 3-37.

72. G. P. Sabol, et al., "Characterization of Corrosion Product Deposits on Fuel in HighTemperature Plants", to be presented at Fontevraud IV, Fontevraud, France,September 1998.

73. Evaluation of Zinc Addition on Fuel Cladding Corrosion at the Halden Test Reactor,Palo Alto, Calif..: Electric Power Research Institute. August 1996. TR-106357.

74. D. Nieder, U. Staudt and B. Stellwag “Zinc Addition for Radiation Field Reduction:Status of and Experience Gathered in German PWRs,” Fontevraud IV InternationalSymposium, Contribution of Materials Investigation to the Resolution of ProblemsEncountered in Pressurized Water Reactors, September 14-18, 1998.

75. C. Wood, "Recommendation for RCS pH Control," Electric Power Research Institute,Letter to PWR Primary Water Chemistry Guideline Committee, June 13, 1997.

76. S. G. Sawochka, Impact of PWR Primary Chemistry on Corrosion ProductDeposition on Fuel Cladding Surfaces, Electric Power Research Institute, Palo Alto,Calif. November 1997. TR-108783.

77. Robust Fuel Program Implementation Plan - Rev. 2, Electric Power ResearchInstitute, March 30, 1998.

78. A. Strasser, et al., Thermal-Hydraulic Bases for Fuel Cycle Designs to Prevent AxialOffset Anomalies, Electric Power Research Institute, Palo Alto, Calif. December1977. TR-108781.

79. Steam Generator Dose Rates on Westinghouse Pressurized Water Reactors. PaloAlto, Calif.: Electric Power Research Institute. June 1982. NP-2453.

References

5-8

80. Effects of Cold Shutdown Chemistry on PWR Radiation Control. Palo Alto, Calif.:Electric Power Research Institute. September 1983. NP-3245.

81. PWR Radiation Fields Through 1982. Palo Alto, Calif.: Electric Power ResearchInstitute. February 1984. NP-3432.

82. R. Roofthooft, et al., Primary Circuit Chemistry of Western PWR and VVER PowerPlants, Unipede report 02004Ren9653, Unipede, Paris, 1996.

83. Evaluation of Cobalt Sources in Westinghouse-Designed Three- and Four-LoopPlants. Palo Alto, Calif.: Electric Power Research Institute. October 1982. NP-2681.

84. The Role of Coolant Chemistry on Plant Radiation Fields. Palo Alto, Calif.: ElectricPower Research Institute. October 1985. NP-4247.

85. Variables Influencing Radiation Fields at Four Pressurized Water Reactors, PaloAlto, Calif.: Electric Power Research Institute. October 1985. NP-4251.

86. M. Hecht, W. C. Schroeder, E. P. Partridge and S. F. Whirl, “Boiler Corrosion,” TheCorrosion Handbook, H. H. Uhlig editor. New York, John Wiley & Sons, Inc. 1948.pg. 520 et seq.

87. W. D. Fletcher, “Primary Coolant Chemistry of PWRs,” Proceedings of 31stInternational Water Conference. Pittsburgh, Pa. Engineers Society of WesternPennsylvania. 1970, pp. 57-66.

88. Corrosion Product Release in Light Water Reactors, Palo Alto, Calif.: Electric PowerResearch Institute. September 1989. NP-6512.

89. D. H. Lister, R. D. Davidson and E. McAlpine, “The Mechanism and Kinetics ofCorrosion Product Release from Stainless Steel in Lithiated High TemperatureWater,” Corrosion Science 27(2), 113 (1987).

90. E. McAlpine, D. H. Lister and H. Ocken, “Corrosion-Induced Release of AlloyConstituents from LWR Materials in High-Temperature Water,” MaterialsPerformance 24, 33 (1985).

91. W. T. Lindsay, Jr. and K. Kussmaul, “Correlation of Iron Solubility from NickelFerrites,” Westinghouse R&D Report 88-8S21-WBPED-R1 (March, 1988).

92. S. M. Walker and E. W. Thornton, “Reanalysis of Oxide Solubility Data: Results forNickel and Cobalt Substituted Ferrites,” Nuclear Electric (UK) ReportTD/RIB/REP/0016, August, 1990.

References

5-9

93. In-Pile Loop Studies of the Effect of Coolant pH on Corrosion Product RadionuclideDeposition, Electric Power Research Institute, Palo Alto, Calif.: February 1992. TR-100156.

94. M. J. Driscoll, G. E. Kohse and O. K. Harling, “Verification Tests of Reliability ofImproved Coolant Chemistry Control Technology,” MIT Nuclear Reactor ReportMITNRL-058, March, 1994.

95. M. V. Polley, “A Correlation Between Operation of Primary Coolants at Low pH(t)with Steam Generator Channel Head Dose Rates in Westinghouse PressurizedWater Reactors, Nuclear Technology, Vol. 71, No. 3, 557-568, 1985.

96. An Evaluation of PWR Radiation Fields, Electric Power Research Institute, PaloAlto, Calif.: April 1992. NP-100306.

97. C. A. Bergmann and J. D. Perock, “Evaluation of Factors Affecting Radiation FieldTrends in Westinghouse-Designed Plants.” Paper No. 3 at the Sixth WaterChemistry of Nuclear Reactor Systems, Bournemouth, England, October 1982.

98. A Survey of the Effect of Primary Coolant pH on Westinghouse PWR PlantRadiation Fields, Electric Power Research Institute, Palo Alto, Calif. November,1994. TR-104180

99. PWR Primary System Chemistry: Experiences with Elevated pH at Millstone PointUnit 3, Electric Power Research Institute, Palo Alto, Calif. July 1992. TR-100960.

100. Operating Limits for Silica, Calcium, Aluminum and Magnesium in PWRs,Electric Power Research Institute, Palo Alto, Calif. TR-107992, September 1997.

101. PWR Shutdown Chemistry Practices, Electric Power Research Institute, Palo Alto,Calif. December 1998. TR-109569.

102. W. A Byers and J. M. Partezana, "The Structure and Chemistry of Deposits onHigh Power PWR Fuel Assemblies," presented at EPRI PWR Chemistry Meeting,Huntington Beach, Calif., Sept. 1-3, 1998.

103. D. R. McCracken, et al., Aspects of the Physics and Chemistry of Water Radiolysisby Fast Neutrons and Fast Electrons in Nuclear Reactors, Report AECL-11895,AECL, Mississauga, Ontario, September 1998.

104. PWR Axial Offset Anomaly Guidelines, Electric Power Research Institute, PaloAlto, Calif., to be published, EPRI TR-110070.

105. G. L. Schultz, "Primary Shutdown Chemistry, Callaway Plant," R. Litman,"Experience with Axial Offset During Fuel Cycles Five and Six at Seabrook Station,"

References

5-10

D. Bryant, "Coordinated pH Control (7.1) with Elevated Lithium at the South TexasProject Electric Generating Station," presented at 1998 EPRI Plant ChemistryMeeting, Huntington Beach, Calif., September 1-3, 1998.

106. Rootcause Investigation of Axial Power Offset Anomaly, Electric Power ResearchInstitute, Palo Alto, Calif., June 1997, EPRI TR-108230.

107. Impact of PWR Primary Chemistry on Corrosion Product Deposition on FuelCladding Surfaces, Electric Power Research Institute, Palo Alto, Calif., November1997, EPRI TR-108783.

A-1

A CALCULATION OF PHT AND DATA EVALUATION

METHODOLOGIES

1.0 Calculation of pHT

To provide a uniform basis for establishing a pHT control program and for comparingobservations at different plants, it is mandatory that industry personnel employ similarexpressions for pertinent chemical equilibria, e.g., the ionization of boric acid. Therecommended relations are given below. Note that in all cases, the pH is defined andshould be reported as the negative logarithm (to the base 10) of the molalconcentrations of the hydrogen ion.

Chemical Expressions, Equilibrium Constants, and Other Definitions forComputing pH Values and Specific Conductivity in PWR Primary Coolant

Definition of pH

pH = -log10 [H+], where [H+] is the molal concentration

Equilibrium Expressions

Kw = aH+ x aOH- = γ+2 x [H+] x [OH-]

KLi = aLi+ x aOH-/aLiOH = γ+2 x [Li+] x [OH-]/[LiOH]

KNH3 = aNH4

+ x aOH-/aNH3 = γ+

2 x [NH4+] x [OH-]/[NH3]

KB1 = aB(OH)4-/aOH- x aB(OH)3

= [B(OH)4-]/[OH-] x [B(OH)3]

KB2 = aB2(OH)7-/aOH- x a2

B(OH)3 = [B2(OH)7

-]/[OH-] x [B(OH)3]2

KB3 = aB3(OH)10-/aOH- x a3

B(OH)3 = [B3(OH)10

-]/[OH-] x [B(OH)3]3

Calculation of PHt and Data Evaluation Methodologies

A-2

KB4 = aB4(OH)14-2/a2

OH- x a4B(OH)3

= γ+2 x [B4(OH)14

-2]/[OH-]2 x [B(OH)3]4

Activity Coefficients

γ+ = Mean ionic activity coefficient for univalent ions. The activitycoefficients for multivalent ions are derived from the univalent ioncoefficients as log10 γ+ (multivalent ion) = zi

2 x log10 γ+ (univalent ion)where zi is the ionic charge of the multivalent ion. The expressionslisted above have taken this relationship into account.

Log10(γ+) = -(DHC x (I.S.)0.5)/(1 + (I.S.)0.5), where I.S., the ionic strength of thesolution = Σi (zi

2 x [Ci])/2. DHC is the Debye-Huckel limiting slope,an empirically-derived expression as a function of temperature, listedin the Table of Constants with the temperature coefficients for theequilibrium constants.

Polynomial Expressions for Equilibrium Constants vs. Temperature

The constants for the polynomial expressions of the equilibrium constants are listedbelow. The polynomial expressions vs. temperature are curve-fitted to informationprovided in the listed references as log10K = f(T), where T is in °K or °C as noted in theTable of Constants.

The equations are expressed in one of two forms:

Form 1: log10K = A/T + B + CT + Dlog10T

Form 2: log10K = A + BT + CT2 + DT3 + ET4

and, DHC = A + BT + CT2 + DT3 + ET4

Definition of Atomic/Molecular Weights

Species Value Reference

Li-7 7.0 grams/mole (5)NH3 17.0 grams/mole (5)

B 10.8 grams/mole (5)


A-3

Table of ConstantsSpecies Charge Expression Form Temp

(°C or °K)Ref.

B(OH)4- -1 KB1 = [B(OH)4-]/[B(OH)3][OH-] 1 °K (1)

A=1573.21B=28.6059 C=.012078D=-13.2258

B2(OH)7- -1 KB2 = [B2(OH)7-]/[B(OH)3]2[OH-] 1 °K (1)

A=2756.1B=-18.996D=5.835

B3(OH)10- -1 KB3 = [B3(OH)10-]/[B(OH)3]3[OH-] 1 °K (1)

A=3339.5B=-8.084D=1.497

B4(OH)14-2 -2 KB4 = γ+2[B4(OH)14-2]/[B(OH)3]4[OH-]2 1 °K (1)

A=12820B=-134.56D=42.105

NH4+ +1 KNH3 = γ+2[NH4+][OH-]/[NH3] 1 °K (2)

A=6730.12B=-264.615C=-.076595D=105.1076

Li+ +1 KLi = γ+2[Li+][OH-]/[LiOH] 2 °C (3)

A=-.7532B=-.0048C=6.746 E-6

H2O 0 Kw = γ+2[H+][OH-] 2 °C (4)

A=-14.9378B=.0424044C=-2.10252 E-4D=6.22026 E-7E=-8.73826 E-10

DHC

(Debye-Huckel Const.

for g+ univalent ion)

N/A Log10γ+= - (DHC(I.S.)0.5)/(1 + (I.S.)0.5) 2 °C (2)

A=0.5027B=-9.028 E-4C=3.315 E-5D=-1.709 E-7E=3.29 E-10


A-4

A graphical summary of the results of the pHT calculation using chemWORKSTM pHCalculator for a core average temperature of 275°C is given in Figure A-1. Tabularresults at temperatures of 275°C to 315°C at 5C° increments are given in Tables A-1through A-9. Results are provided for lithium ranges yielding at most 1-1/2% errorwith respect to exact equations.

2.0 Data Consistency (pH at 25°C/Conductivity at 25°C/Boron/Lithium)

Because of the importance of primary chemistry control relative to system integrity,fuel integrity, activity buildup and plant availability, it is mandatory that a high levelof assurance be provided by the Chemistry Department that all water chemistryparameters remain well within specifications. This requires that accurate and timelychemistry results are generated by the chemistry technicians, that errors in thechemistry data are identified and resolved promptly, and that results are evaluatedthoroughly to assess system status. In addition, a prompt indication of the presence ofcontaminants for which routine analyses are not performed can be of significant valuerelative to problem avoidance. Such an indication can be a valid technical basis forreducing the sampling frequencies of some difficult-to-measure species. Fortunately,the consistency of the pH and conductivity data at 25°C with the boron and lithiumconcentrations can be used as the indicator of the presence of impurities.

Specifically, in the absence of significant concentrations of impurities in the primarycoolant, a well-defined relation exists between the pH, specific conductivity, lithiumand boron concentrations at 25°C. The dependence of pH on lithium and boronconcentrations at 25°C is given in Table A-10 (calculated using EPRI chemWORKSTM

pH Calculator).

Approximate information for the conductivity/lithium/boron relation is given in TableA-11 (calculated using EPRI chemWORKSTM pH Calculator). Unfortunately, values ofthe equivalent conductances of the major anionic boron species (B(OH)4

- and B3 (OH)10-)

estimated from data developed by different investigators vary significantly. The valuesused in developing Table A-11 are 40 mho-cm2/equivalent for B(OH)4

- and 27 mho-cm2/equivalent for B3 (OH) 10

- (see Table A-12).

Using such information, the relation of conductivity and pH to the lithium and boronconcentrations can be used to verify the consistency of the data array. For example, at1.0 ppm lithium and 300 ppm boron, the pH should be 6.9 and the conductivityapproximately 11 µS/cm. If the measured pH deviates by more than approximately 0.1unit from 6.9 or the conductivity deviates by more than 10% (or in this case 1 µS/cm)one of the two following conclusions can be reached:

1. One or both of the measurements is in error. Review of trend graphs for eachparameter can sometimes give an indication of the parameter in error.


A-5

2. Presuming the analytical results are correct, an unidentified species is present in thecoolant. For example, a neutral salt (e.g., sodium chloride) will increase solutionconductivity and not affect the pH. The presence of a base will increase theconductivity and pH values and an acid will increase the conductivity but depressthe pH. If the observed conductivity is lower than calculated, a measurement errorshould be expected.

As noted above, this data consistency check can assist in demonstrating the presence ofa contaminant and also can provide insights into the nature of the contaminant.However, caution should be applied as acceptable analytical uncertainties may bemisinterpreted as indications of contaminants when high concentrations of boron andlithium are present. For example, a solution of only 2.20 ppm Li/1300 ppm B will havea conductivity of 22.94 µS/cm and a pH25°C of 6.29. The same solution with 0.15 ppm Cland 0.15 ppm F will have a conductivity of 23.21 µS/cm and a pH25°C of 6.27. Thesedifferences are within acceptable analytical uncertainty.

3.0 Verification and Validation of the EPRI chemWORKSTM Primary SystempH Calculator

The EPRI chemWORKSTM Primary System pH Calculator calculates the at-temperaturepH given the boron concentration, the lithium concentration, and the ammoniaconcentration. It also calculates any of these three parameters if two of them are knownand the pHT is given. To resolve slight differences between values calculated using theEPRI chemWORKSTM pH Calculator and some data found in the literature, EPRIperformed investigations to check the accuracy of the software. The investigationswere conducted as a verification and validation (V/V) of the results produced by thesoftware and the results developed directly using the equations given in Section 1.0 ofthis appendix.

First, the correlations for the temperature dependence of thermodynamic data used inthe EPRI chemWORKSTM calculations were compared to the original data from theliterature (see the references at the end of this appendix). The values depending on thetemperature are the ion product of water Kw, the boric acid equilibrium constants KB1 to 4,the lithium hydroxide equilibrium constant KLi, and the Debye-Huckel limiting slopeDHC (which determines the temperature dependence of the activity coefficient of thesolution γ), all given in Section 1.0 of this appendix. It was found that the relationsused to correlate Kw data by EPRI chemWORKSTM were consistent with the dataavailable up to 316°C in the references. For the DHC where original data are availableat higher temperatures, consistency was demonstrated up to 350°C. For KB1 to 4 , thetemperature functionality of the coefficients written in this appendix were examinedagainst the data of Mesmer et al. (6) over the temperature range of 285°C to 310°C andfound to be consistent with those data. The KLi correlation written in this appendix isreproduced identically in the Calculator code. Consequently, it was concluded that the


A-6

thermodynamic data used by the EPRI chemWORKSTM pH Calculator are accurateunder primary system conditions.

Second, the results produced by the EPRI chemWORKSTM pH Calculator werereviewed for three temperature cases, 285°C, 300°C and 310°C, over a range of pHT

from 6.9 to 7.4 and a range of boron concentration from 100 to 1500 ppm. The pHCalculator was used in the lithium mode, i.e., the mode for which the inputs are thepH, the boron concentration, and the ammonia concentration (0 in this case), and theoutput is the lithium concentration. The accuracy of the EPRI chemWORKSTM pHCalculator was reviewed against independent calculations employing the exactequations and the data set forth in Section 1.0 of this appendix. The Calculator outputwas found to be within a band of 0.25% of the independent calculations over the entirerange of parameters under investigation. It was therefore concluded that the EPRIchemWORKSTM pH Calculator is reliable and accurate.

Tables A-1 through A-9 were calculated using the EPRI chemWORKSTM pHCalculator. In all cases except those entries italicized, the difference between theCalculator output and entry in the analogous table of Revision 3.0 of these Guidelines isless than 1.5%. The difference for italicized entries is between 1.5% and 2.0%.


A-7

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

0200400600800100012001400160018002000

Boron concentration, ppm

Lith

ium

con

cent

ratio

n, p

pm

pH = 6.7

pH = 6.8

pH = 6.9

pH = 7.0 pH = 7.1 pH = 7.2 pH = 7.3 pH = 7.4

Figure A-1Lithium-Boron Relationships for Various pH at 275°C


A-8

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.21 0.26 0.33 0.42 0.53 0.68 0.86 1.09 1.3850 0.28 0.36 0.45 0.57 0.72 0.91 1.16 1.47 1.87100 0.36 0.45 0.57 0.72 0.92 1.16 1.47 1.87 2.38150 0.43 0.55 0.69 0.88 1.11 1.41 1.79 2.28 2.90200 0.51 0.65 0.82 1.04 1.31 1.67 2.12 2.69 3.43250 0.59 0.75 0.94 1.20 1.52 1.93 2.45 3.12 3.98300 0.67 0.85 1.07 1.36 1.73 2.20 2.79 3.55400 0.83 1.06 1.34 1.70 2.16 2.75 3.50500 1.00 1.27 1.62 2.05 2.61 3.32600 1.18 1.50 1.90 2.42 3.08 3.92700 1.36 1.73 2.20 2.80 3.56800 1.55 1.97 2.51 3.19 4.07900 1.74 2.22 2.82 3.59

1000 1.94 2.48 3.15 4.021100 2.16 2.74 3.491200 2.37 3.01 3.841300 2.59 3.301400 2.82 3.591500 3.05 3.901600 3.301700 3.551800 3.801900 4.082000

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.19 0.25 0.31 0.40 0.50 0.64 0.80 1.02 1.3050 0.26 0.33 0.42 0.53 0.67 0.85 1.08 1.37 1.74100 0.33 0.42 0.53 0.67 0.85 1.07 1.36 1.73 2.20150 0.40 0.50 0.64 0.81 1.02 1.30 1.65 2.10 2.67200 0.47 0.59 0.75 0.95 1.21 1.53 1.94 2.47 3.15250 0.54 0.68 0.87 1.10 1.39 1.77 2.25 2.86 3.64300 0.61 0.77 0.98 1.24 1.58 2.01 2.56 3.25400 0.76 0.96 1.22 1.55 1.97 2.51 3.19 4.07500 0.91 1.16 1.47 1.87 2.38 3.02 3.86600 1.07 1.36 1.73 2.20 2.79 3.56700 1.23 1.57 1.99 2.54 3.23800 1.41 1.78 2.27 2.89 3.68900 1.58 2.01 2.56 3.25

1000 1.76 2.24 2.85 3.631100 1.94 2.48 3.15 4.021200 2.14 2.72 3.471300 2.34 2.98 3.791400 2.54 3.241500 2.75 3.511600 2.97 3.781700 3.191800 3.421900 3.662000 3.91

Relation of pH at 280 C to Lithium and Boron Concentrations

Li, ppm

Table A-1Relation of pH at 275 C to Lithium and Boron Concentrations

Li, ppm

Table A-2


A-9

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.18 0.23 0.29 0.37 0.47 0.59 0.75 0.95 1.2050 0.24 0.30 0.39 0.49 0.62 0.78 0.99 1.26 1.60

100 0.30 0.38 0.48 0.61 0.77 0.98 1.24 1.58 2.01150 0.36 0.46 0.58 0.74 0.93 1.18 1.50 1.91 2.43200 0.42 0.54 0.68 0.87 1.10 1.39 1.77 2.25 2.86250 0.49 0.62 0.78 1.00 1.26 1.60 2.04 2.59 3.30300 0.55 0.70 0.89 1.13 1.43 1.82 2.31 2.94 3.75400 0.69 0.87 1.10 1.40 1.78 2.26 2.88 3.67500 0.82 1.04 1.33 1.68 2.14 2.72 3.47600 0.96 1.22 1.55 1.97 2.51 3.20 4.09700 1.11 1.41 1.79 2.28 2.90 3.69800 1.26 1.60 2.04 2.59 3.30900 1.42 1.80 2.29 2.92 3.71

1000 1.57 2.01 2.55 3.251100 1.74 2.22 2.82 3.591200 1.91 2.44 3.10 3.961300 2.09 2.66 3.391400 2.27 2.89 3.681500 2.46 3.13 4.001600 2.65 3.381700 2.84 3.631800 3.05 3.901900 3.262000 3.48

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.17 0.21 0.27 0.34 0.43 0.54 0.69 0.87 1.1150 0.22 0.28 0.35 0.44 0.56 0.71 0.91 1.15 1.46

100 0.27 0.35 0.44 0.55 0.70 0.89 1.13 1.43 1.82150 0.33 0.41 0.52 0.67 0.84 1.07 1.36 1.72 2.19200 0.38 0.48 0.61 0.78 0.99 1.25 1.59 2.02 2.57250 0.44 0.56 0.70 0.89 1.13 1.44 1.83 2.33 2.96300 0.50 0.63 0.80 1.01 1.28 1.63 2.07 2.64 3.36400 0.62 0.78 0.99 1.25 1.59 2.02 2.57 3.27500 0.73 0.93 1.18 1.50 1.90 2.43 3.09 3.95600 0.86 1.09 1.38 1.76 2.24 2.85 3.63700 0.99 1.25 1.59 2.03 2.58 3.28800 1.12 1.42 1.81 2.30 2.93 3.73900 1.25 1.60 2.03 2.59 3.30

1000 1.40 1.78 2.26 2.88 3.671100 1.54 1.96 2.50 3.18 4.071200 1.69 2.16 2.75 3.501300 1.85 2.35 3.00 3.831400 2.01 2.56 3.261500 2.17 2.77 3.521600 2.34 2.98 3.801700 2.51 3.20 4.091800 2.69 3.431900 2.88 3.672000 3.07 3.92

Table A-3

Li, ppm


Li, ppm



A-10

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.15 0.19 0.24 0.31 0.39 0.49 0.62 0.79 1.0050 0.20 0.25 0.32 0.40 0.51 0.64 0.82 1.03 1.31100 0.24 0.31 0.39 0.49 0.63 0.80 1.01 1.28 1.63150 0.29 0.37 0.47 0.59 0.75 0.95 1.21 1.54 1.95200 0.34 0.43 0.55 0.69 0.88 1.11 1.41 1.79 2.29250 0.39 0.49 0.63 0.79 1.01 1.27 1.62 2.06 2.63300 0.44 0.56 0.71 0.90 1.14 1.44 1.83 2.33 2.97400 0.54 0.69 0.87 1.11 1.40 1.78 2.27 2.89 3.68500 0.65 0.82 1.04 1.32 1.68 2.14 2.72 3.47600 0.76 0.96 1.22 1.55 1.96 2.51 3.19700 0.87 1.10 1.40 1.78 2.27 2.89 3.68800 0.98 1.25 1.59 2.02 2.57 3.28900 1.10 1.40 1.78 2.27 2.89 3.68

1000 1.22 1.56 1.98 2.53 3.221100 1.35 1.72 2.19 2.79 3.561200 1.48 1.89 2.40 3.06 3.911300 1.62 2.06 2.62 3.341400 1.75 2.24 2.85 3.631500 1.89 2.42 3.08 3.941600 2.04 2.61 3.321700 2.19 2.80 3.571800 2.35 3.00 3.831900 2.51 3.20 4.092000 2.67 3.41

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.13 0.17 0.22 0.27 0.35 0.44 0.56 0.71 0.9050 0.18 0.22 0.28 0.36 0.45 0.57 0.72 0.92 1.16100 0.22 0.27 0.35 0.44 0.55 0.70 0.89 1.13 1.44150 0.26 0.32 0.41 0.52 0.66 0.84 1.06 1.35 1.72200 0.30 0.38 0.48 0.61 0.77 0.98 1.24 1.58 2.01250 0.34 0.43 0.55 0.69 0.88 1.12 1.42 1.80 2.30300 0.38 0.49 0.62 0.78 0.99 1.26 1.60 2.04 2.60400 0.47 0.60 0.76 0.96 1.22 1.56 1.98 2.52 3.22500 0.56 0.72 0.91 1.15 1.46 1.86 2.37 3.02 3.86600 0.66 0.84 1.06 1.35 1.71 2.18 2.78 3.54700 0.76 0.96 1.22 1.55 1.96 2.51 3.20800 0.86 1.09 1.38 1.75 2.23 2.85 3.63900 0.96 1.22 1.55 1.96 2.51 3.19 4.09

1000 1.06 1.35 1.72 2.19 2.79 3.551100 1.17 1.49 1.89 2.42 3.08 3.941200 1.28 1.63 2.08 2.65 3.381300 1.40 1.78 2.27 2.89 3.681400 1.52 1.93 2.46 3.14 4.011500 1.64 2.09 2.66 3.391600 1.76 2.25 2.87 3.651700 1.89 2.41 3.08 3.931800 2.03 2.58 3.291900 2.16 2.76 3.522000 2.30 2.94 3.74


Li, ppm


Li, ppm


A-11

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.12 0.15 0.19 0.24 0.31 0.39 0.49 0.62 0.7950 0.15 0.19 0.25 0.31 0.39 0.50 0.63 0.80 1.02100 0.19 0.24 0.30 0.38 0.48 0.61 0.78 0.99 1.25150 0.22 0.28 0.36 0.45 0.57 0.73 0.92 1.17 1.49200 0.26 0.33 0.42 0.53 0.67 0.85 1.07 1.37 1.74250 0.30 0.37 0.47 0.60 0.76 0.97 1.23 1.56 1.99300 0.33 0.42 0.53 0.68 0.86 1.09 1.38 1.76 2.24400 0.41 0.52 0.66 0.83 1.05 1.34 1.70 2.17 2.76500 0.48 0.62 0.78 0.99 1.26 1.60 2.04 2.60 3.31600 0.57 0.72 0.91 1.16 1.47 1.87 2.38 3.03 3.89700 0.65 0.82 1.04 1.33 1.68 2.15 2.74 3.49800 0.73 0.93 1.18 1.50 1.91 2.44 3.10 3.97900 0.82 1.04 1.32 1.68 2.14 2.73 3.49

1000 0.91 1.16 1.47 1.87 2.38 3.04 3.891100 1.00 1.27 1.62 2.06 2.63 3.351200 1.10 1.39 1.77 2.26 2.88 3.681300 1.19 1.52 1.93 2.46 3.14 4.021400 1.30 1.64 2.10 2.67 3.411500 1.40 1.78 2.27 2.89 3.691600 1.50 1.91 2.44 3.11 3.981700 1.61 2.06 2.62 3.341800 1.72 2.20 2.80 3.581900 1.84 2.34 2.99 3.832000 1.95 2.49 3.18

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.10 0.13 0.17 0.21 0.27 0.34 0.43 0.54 0.6950 0.13 0.17 0.21 0.27 0.34 0.43 0.55 0.69 0.88100 0.16 0.20 0.26 0.33 0.42 0.53 0.67 0.85 1.07150 0.19 0.24 0.31 0.39 0.49 0.63 0.79 1.01 1.27200 0.22 0.28 0.36 0.45 0.57 0.72 0.92 1.17 1.48250 0.25 0.32 0.41 0.51 0.65 0.83 1.05 1.33 1.69300 0.28 0.36 0.45 0.58 0.73 0.93 1.18 1.50 1.90400 0.35 0.44 0.56 0.71 0.90 1.14 1.45 1.84 2.35500 0.41 0.52 0.66 0.84 1.07 1.36 1.73 2.20 2.81600 0.48 0.61 0.77 0.98 1.24 1.58 2.02 2.57 3.28700 0.55 0.70 0.88 1.12 1.43 1.81 2.32 2.95 3.78800 0.62 0.79 1.00 1.27 1.62 2.06 2.62 3.35900 0.69 0.88 1.12 1.42 1.81 2.31 2.94 3.76

1000 0.77 0.98 1.24 1.58 2.01 2.56 3.271100 0.85 1.07 1.37 1.74 2.22 2.83 3.611200 0.93 1.18 1.50 1.90 2.43 3.10 3.971300 1.01 1.28 1.63 2.08 2.65 3.381400 1.09 1.39 1.76 2.25 2.87 3.671500 1.18 1.50 1.90 2.43 3.10 3.971600 1.26 1.61 2.05 2.62 3.341700 1.36 1.73 2.20 2.81 3.581800 1.45 1.84 2.35 3.00 3.841900 1.55 1.96 2.51 3.202000 1.64 2.09 2.67 3.41

Li, ppm


Li, ppm


Table A-7


A-12

pHt 6.70 6.80 6.90 7.00 7.10 7.20 7.30 7.40 7.50B, ppm

0 0.09 0.11 0.14 0.18 0.23 0.29 0.36 0.46 0.5950 0.11 0.14 0.18 0.23 0.29 0.37 0.46 0.59 0.75100 0.14 0.17 0.22 0.28 0.35 0.45 0.56 0.72 0.91150 0.16 0.20 0.26 0.33 0.42 0.53 0.67 0.85 1.07200 0.19 0.24 0.30 0.38 0.48 0.61 0.77 0.98 1.25250 0.21 0.27 0.34 0.43 0.55 0.69 0.88 1.12 1.42300 0.24 0.30 0.38 0.48 0.61 0.78 0.99 1.25 1.60400 0.29 0.37 0.47 0.59 0.75 0.95 1.21 1.54 1.96500 0.34 0.44 0.55 0.70 0.89 1.13 1.44 1.83 2.34600 0.40 0.51 0.65 0.82 1.04 1.32 1.68 2.14 2.73700 0.46 0.58 0.74 0.93 1.19 1.51 1.92 2.46 3.14800 0.52 0.66 0.83 1.06 1.35 1.71 2.18 2.79 3.56900 0.58 0.73 0.93 1.18 1.51 1.91 2.45 3.12 4.00

1000 0.64 0.81 1.03 1.31 1.67 2.13 2.72 3.471100 0.70 0.89 1.14 1.45 1.84 2.35 3.00 3.841200 0.77 0.98 1.24 1.58 2.02 2.57 3.281300 0.84 1.06 1.35 1.72 2.20 2.80 3.581400 0.91 1.15 1.47 1.86 2.38 3.04 3.891500 0.98 1.24 1.58 2.02 2.57 3.281600 1.05 1.34 1.70 2.17 2.77 3.531700 1.12 1.43 1.82 2.32 2.97 3.801800 1.20 1.53 1.94 2.48 3.171900 1.28 1.63 2.08 2.65 3.382000 1.36 1.73 2.21 2.82 3.60


Li, ppm


A-13

Table A-10pH of Boric Acid - Lithium Hydroxide Solutions at 25°C

B (ppm) 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.000 7.00 9.55 9.85 10.02 10.14 10.24 10.32 10.38 10.44 10.49 10.53 10.57 10.61 10.64 10.68 10.70 10.73

50 5.79 7.12 7.43 7.60 7.73 7.83 7.91 7.98 8.04 8.09 8.14 8.19 8.23 8.26 8.30 8.33 8.36

100 5.63 6.82 7.12 7.29 7.42 7.52 7.59 7.66 7.72 7.77 7.82 7.86 7.90 7.94 7.97 8.00 8.03

150 5.54 6.64 6.93 7.11 7.23 7.33 7.41 7.48 7.53 7.59 7.63 7.67 7.71 7.75 7.78 7.81 7.84

200 5.47 6.50 6.80 6.97 7.10 7.20 7.27 7.34 7.40 7.45 7.50 7.54 7.57 7.61 7.64 7.67 7.70

250 5.42 6.40 6.69 6.87 6.99 7.09 7.17 7.23 7.29 7.34 7.39 7.43 7.47 7.50 7.53 7.56 7.59

300 5.38 6.31 6.60 6.78 6.90 7.00 7.07 7.14 7.20 7.25 7.29 7.34 7.37 7.41 7.44 7.47 7.50

400 5.30 6.15 6.45 6.62 6.74 6.84 6.92 6.99 7.04 7.09 7.14 7.18 7.22 7.25 7.28 7.31 7.34

500 5.23 6.03 6.32 6.49 6.61 6.71 6.79 6.85 6.91 6.96 7.01 7.05 7.09 7.12 7.15 7.18 7.21

600 5.17 5.91 6.20 6.37 6.50 6.59 6.67 6.74 6.79 6.85 6.89 6.93 6.97 7.00 7.04 7.07 7.09

700 5.12 5.81 6.09 6.27 6.39 6.49 6.56 6.63 6.69 6.74 6.78 6.82 6.86 6.90 6.93 6.96 6.99

800 5.07 5.71 6.00 6.17 6.29 6.39 6.46 6.53 6.59 6.64 6.68 6.72 6.76 6.80 6.83 6.86 6.89

900 5.02 5.62 5.90 6.07 6.20 6.29 6.37 6.44 6.49 6.54 6.59 6.63 6.67 6.70 6.73 6.76 6.79

1000 4.98 5.54 5.82 5.99 6.11 6.20 6.28 6.35 6.40 6.46 6.50 6.54 6.58 6.61 6.65 6.68 6.70

1100 4.94 5.46 5.73 5.90 6.02 6.12 6.20 6.26 6.32 6.37 6.42 6.46 6.49 6.53 6.56 6.59 6.62

1200 4.90 5.39 5.66 5.82 5.94 6.04 6.12 6.18 6.24 6.29 6.34 6.38 6.41 6.45 6.48 6.51 6.54

1300 4.86 5.32 5.58 5.75 5.87 5.96 6.04 6.11 6.16 6.21 6.26 6.30 6.34 6.37 6.40 6.43 6.46

1400 4.82 5.26 5.51 5.68 5.80 5.89 5.97 6.03 6.09 6.14 6.19 6.23 6.26 6.30 6.33 6.36 6.39

1500 4.79 5.19 5.44 5.61 5.73 5.82 5.90 5.96 6.02 6.07 6.12 6.16 6.19 6.23 6.26 6.29 6.32

1600 4.75 5.14 5.38 5.54 5.66 5.75 5.83 5.90 5.95 6.00 6.05 6.09 6.13 6.16 6.19 6.22 6.25

1700 4.72 5.08 5.32 5.48 5.60 5.69 5.77 5.83 5.89 5.94 5.99 6.03 6.06 6.10 6.13 6.16 6.19

1800 4.69 5.03 5.26 5.42 5.54 5.63 5.71 5.77 5.83 5.88 5.92 5.96 6.00 6.04 6.07 6.10 6.13

1900 4.66 4.98 5.21 5.36 5.48 5.57 5.65 5.71 5.77 5.82 5.86 5.91 5.94 5.98 6.01 6.04 6.07

2000 4.63 4.94 5.15 5.31 5.42 5.51 5.59 5.66 5.71 5.76 5.81 5.85 5.89 5.92 5.95 5.98 6.01


A-14

Table A-11Conductivity of Boric Acid - Lithium Hydroxide Solutions at 25°C

Lithium (ppm Li)

B (ppm) 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00

0 0.1 8.5 16.9 25.4 33.8 42.3 50.7 59.1 67.6 76.0 84.4 92.8 >100 >100 >100 >100 >100

50 0.6 2.9 5.7 8.5 11.3 14.2 17.0 19.8 22.6 25.5 28.3 31.1 34.0 36.8 39.6 42.5 45.3

100 0.9 2.9 5.7 8.5 11.3 14.1 16.9 19.7 22.5 25.3 28.1 31.0 33.8 36.6 39.4 42.2 45.0

150 1.1 2.9 5.7 8.4 11.2 14.0 16.8 19.6 22.4 25.2 28.0 30.8 33.6 36.5 39.3 42.1 44.9

200 1.3 2.9 5.6 8.4 11.2 14.0 16.8 19.6 22.4 25.1 27.9 30.7 33.5 36.3 39.1 41.9 44.7

250 1.5 2.9 5.6 8.4 11.2 13.9 16.7 19.5 22.3 25.0 27.8 30.6 33.4 36.2 38.9 41.7 44.5

300 1.6 3.0 5.6 8.4 11.1 13.9 16.6 19.4 22.2 24.9 27.7 30.5 33.2 36.0 38.8 41.5 44.3

400 2.0 3.0 5.6 8.3 11.0 13.8 16.5 19.2 22.0 24.7 27.4 30.2 32.9 35.6 38.4 41.1 43.9

500 2.3 3.1 5.6 8.3 10.9 13.6 16.3 19.0 21.7 24.4 27.1 29.8 32.6 35.3 38.0 40.7 43.4

600 2.6 3.2 5.6 8.2 10.8 13.5 16.2 18.8 21.5 24.2 26.8 29.5 32.2 34.9 37.6 40.2 42.9

700 2.9 3.2 5.6 8.2 10.8 13.4 16.0 18.6 21.3 23.9 26.6 29.2 31.9 34.5 37.2 39.8 42.5

800 3.3 3.4 5.6 8.1 10.7 13.3 15.9 18.5 21.1 23.7 26.3 28.9 31.5 34.2 36.8 39.4 42.0

900 3.6 3.5 5.7 8.1 10.6 13.2 15.7 18.3 20.9 23.5 26.1 28.7 31.2 33.8 36.4 39.0 41.6

1000 4.0 3.7 5.7 8.1 10.6 13.1 15.6 18.2 20.7 23.3 25.8 28.4 31.0 33.5 36.1 38.7 41.3

1100 4.4 3.9 5.8 8.1 10.6 13.0 15.5 18.1 20.6 23.1 25.7 28.2 30.7 33.3 35.8 38.4 40.9

1200 4.9 4.1 5.9 8.2 10.6 13.0 15.5 18.0 20.5 23.0 25.5 28.0 30.5 33.1 35.6 38.1 40.6

1300 5.3 4.3 6.0 8.2 10.6 13.0 15.4 17.9 20.4 22.9 25.4 27.9 30.4 32.9 35.4 37.9 40.4

1400 5.8 4.6 6.2 8.3 10.6 13.0 15.4 17.8 20.3 22.8 25.2 27.7 30.2 32.7 35.2 37.7 40.2

1500 6.2 4.9 6.3 8.4 10.7 13.0 15.4 17.8 20.3 22.7 25.2 27.6 30.1 32.6 35.0 37.5 40.0

1600 6.7 5.3 6.5 8.5 10.7 13.0 15.4 17.8 20.2 22.6 25.1 27.5 30.0 32.4 34.9 37.3 39.8

1700 7.2 5.6 6.7 8.7 10.8 13.1 15.4 17.8 20.2 22.6 25.0 27.5 29.9 32.3 34.8 37.2 39.7

1800 7.8 6.0 7.0 8.8 10.9 13.2 15.5 17.8 20.2 22.6 25.0 27.4 29.8 32.3 34.7 37.1 39.6

1900 8.3 6.4 7.3 9.0 11.0 13.2 15.5 17.9 20.2 22.6 25.0 27.4 29.8 32.2 34.6 37.0 39.5

2000 8.9 6.8 7.5 9.2 11.2 13.3 15.6 17.9 20.2 22.6 25.0 27.4 29.7 32.2 34.6 37.0 39.4


A-15

Table A-12Equivalent Conductance of 25°C

Value Reference

ΛH+ = 350 mho-cm2/equiv. (5)ΛOH-

= 199 mho-cm2/equiv. (5)ΛNH4+ = 74 mho-cm2/equiv. (5)

ΛLi+= 39 mho-cm2/equiv. (5)

ΛB(OH)4- = 40 mho-cm2/equiv. (7)

ΛB2(OH)7- = 34 mho-cm2/equiv. (estimated)

ΛB3(OH)10- = 27 mho-cm2/equiv. (7)

ΛB4(OH)14-2 = 22 mho-cm2/equiv. (estimated)

ΛHSiO3-1 = 40-50 mho-cm2/equiv. (estimated)

Calculation of Equivalent Conductance of 25°C

Specific conductivity at 25°C is calculated by the following expression:

k (µS/cm) = ΣΛi x Ci x 1000

Where:

Λi = The equivalent conductance of species i (mho-cm2/equiv.)Ci = The concentration of species i (equivalents/kg)

4.0 Sampling Considerations for Monitoring RCS Particulates

PWRs do not have a standard requirement to sample the reactor coolant system forcrud analyses during operation, other than that implied by the Standard TechnicalSpecifications forE , the average disintegration energy. Plant specific samplingpractices have evolved in the industry, depending on manpower and interest insampling the reactor coolant for crud activity. Recognizing that major systemmodifications to improve sampling technology are impractical in most existing plants,the purpose of this section is to provide an historical review of sampling technologyliterature which illustrates potential caveats inherent in quantitative interpretation ofdata obtained using existing practices. Though existing practices may not be ideal forindependently following and analyzing soluble versus particulate crud transients in theRCS, it is clear that they are capable of qualitatively trending coolant activity in totaland are useful in detecting and following abnormal releases from the core.


A-16

4.1 Background on RCS Sampling Technology

Two sample points are used for reactor coolant sample analysis in Westinghouse PWRs:the hot leg and the CVCS (letdown), although the cold leg may be available in someplants. B&W PWRs use a cold leg for the reactor coolant sample. The hot leg (or coldleg in B&W PWRs) sample usually (but not always) has a delay coil to provide a 60-second or longer decay of nitrogen-16. The sample line, which is typically 600 feet orlonger of 3/8-inch tubing, is routed through containment to a sample cooler in a centralsample station. Double isolation valves must be opened to initiate sample flow. A 0.1to 0.5-gpm purge to the volume control tank (VCT) for 10 to 15 minutes may be usedfor flushing of the lines; some plants continually purge the sample line to the VCT.Limitations on the number of operating cycles on hot leg isolation valves has promptedsome plant operators to limit the use of hot leg samples, meaning that letdown streamsare the only alternative for many plants. After purging, the local sample valve isopened and the cooled sample is collected for analysis.

The letdown stream at typically 45 to 120 gpm is normally taken from a cold legthrough a delay line and cooled to approximately 290

oF by passage on the shell side of

a regenerative heat exchanger. Pressure reduction valves or orifice plates reduce thepressure to 300 to 350 psig and control the flow of reactor coolant leaving the reactorloop. The letdown heat exchanger cools the stream to approximately 115

oF. The

sample line is typically 175 feet. A concern with letdown samples is crud depositionfrom the pressure reductions and the solubility changes as a result of cooling theletdown stream. The corrosion products are more soluble in the cooled sample streamthan in the reactor coolant.

The installed sampling systems in PWRs were not designed to obtain representativesamples of trace corrosion products; the design criterion was to deliver a representativeliquid sample for soluble constituents without exposing the operator to excessiveradiation. The sample lines are long, narrow-bore tubing and have an uneventemperature gradient over their length. Particle interactions with the sample line wallsand changes in solubility characteristics as the stream is cooled in the boric acid matrixcan preclude obtaining samples sufficiently representative of the actual coolant state fora quantitative interpretation of changes in particulate crud chemistry and transportfrom the core/plant surfaces.

Modeling of sample systems shows that deposits on sample lines build up to a steady-state value quickly at high velocities and slowly at low velocities. Approximately onemonth will be required for equilibrium to be established for 5-micron particlestransported at 121

oC by 0.65-inch ID tubing (8). A 15-minute sample line purge,

therefore, does not establish equilibrium, nor does using a criterion to flush the line forthree sample line volumes at typical sample line flow rates.


A-17

The minimum deposit weight of 5-micron particles on the 0.65 inch ID sample line at121

oC occurs at 6 ft/s linear flow. The linear flow rate through 3/8-inch OD tubing with

a nominal inside diameter of 0.245 inch, however, is only 0.88 ft/s at 0.13 gpm and 3.4ft/s at 0.5 gpm (common purge rates used) rather than 6 ft/s, which would require aflow rate of 0.89 gpm for 3/8-inch tubing. The mass of deposit continually increaseswith an increase in particle diameter up to approximately 1 micron, at which time thedeposited mass stabilizes..

To ensure that the particle concentration in the sample is representative, ASTM D 3370-95a, Standard Practices for Sampling Water from Closed Conduits, recommends sample linestability in addition to linear flow rate considerations. Experience at Sizewell B (9)affirms the ASTM concern for flow stability: the sample flow is stabilized forapproximately two weeks prior to a sampling campaign, and any disturbance (e.g.,valve manipulations) invalidates the results for several days. Maintaining acontinuous, constant flow in the sample line avoids both particulate and solubletransients as a result of the temperature front along the sample line. At Sizewell B thehot leg sample stream to a gross gamma monitor and a boronometer is continuallypurged at about 900 mL/min to the VCT, and a capillary stream from the hot legcontinuously flows at 84 mL/min. Isokinetic sampling is preferred because of varyingsizes and densities of particles in a flowing stream. The UKAEA preference is to placethe capillary head immediately after a sample cooler, but each capillary line is installedimmediately in front of the cooler and cooled individually. Sizewell B experienceindicates that particulates are of lesser importance, except during transients, since theyamount to only a fraction of the total corrosion product burden in the RCS. Theparticulates in the RCS are mainly (<95%) 1 micron or less in diameter and are readilycollected by the sample lines, thus isokinetic sampling is of less importance in theSizewell B work.

A concern in addition to sample delivery considerations and chemistry changesassociated with sample conditioning is both dissolution and precipitation that can occurwhen a liquid sample at the terminal sample point is in contact with oxygen (air) (10).The cooled stream is less basic (or more acidic if lithium is not at sufficient levels) as aresult of the increased overall ionization of boric acid that occurs with temperaturereduction. Both iron and nickel solubility increases with temperature reduction.However, upon oxygenation, nickel solubility can significantly increase, but ironsolubility will significantly decrease upon formation of iron(III) hydroxide. Even inslightly acidic solutions iron(III) forms insoluble hydroxides, co-precipitating otheractivities such as chromium-51. Thus, local grab samples, both hot leg and letdownsamples, can provide misleading information as to the composition of the particulatecrud if the sample is filtered in the laboratory, notwithstanding the fact that the totalconcentration of soluble plus insoluble corrosion product (obtained upon acidificationof the total sample) might qualitatively trend with observed activity transport. Workreported by Bridle (11) used a continually flowing capillary steam approximating


A-18

isokinetic sampling from the hot leg with approximately 112 seconds transient.Particulates were collected on a 0.1-micron Millipore or Nuclepore polycarbonatemembrane filter followed by two Acropor cation membranes and two anionmembranes in the same sample holder. Noe et al. (12) showed that differences insoluble and particulate cobalt-58 and cobalt-60 were related to the sample line history.At high boron at beginning of a fuel cycle, cobalt activity from previous operationsdissolved during delivery in the sample line, resulting a decrease in the soluble cobalt-58 to cobalt-60 activity ratio. This suggests that hot sampling is preferred over samplesthat are conditioned. However, Bridle et al. (13) reported that soluble transition metalions interact with the hot sample line. The pH of the cooled sample is sensitive to theboron/lithium ratio during the cycle, and at end of core life significant pH increasesoccur in the conditioned sample.

Polley and Brookes (14) reported that increasing sample flow rate from 25 g/sec to 100– 125 g/sec resulted in up to a factor of 100 decrease in soluble cobalt-58 and cobalt-60at Ringhals 3. Transients occurred in the sample line from valve operation for up to aweek. Elemental concentrations for the DIDO Water Reactor at Harwell were shown tovary inversely with sample flow rate, particularly for sample lines with long residencetimes. Roesmer has reported studies (15, 16) suggesting “hot” samples are moreindicative of actual crud conditions in the circulating reactor coolant since solubilityincreases with a decrease in temperature. An important consideration in Roesmer’swork was the relationship between the volume of sample filtered and the deposit onthe inline crud filter; large sample volumes resulted in significant dissolution of crudcollected on the inline filter. Dissolution of the crud was particularly important forsamples that were conditioned. Roesmer concluded that the ratio of “hot” to ambientcrud should be between 2.4 and 5.3 if no dissolution from the filter matrix were to takeplace.

4.2 Existing Sampling Practices

The existing sampling systems in domestic PWRs can not be expected to deliverrepresentative samples of distinct particulate vs. colloidal vs. dissolved corrosionproducts as presently configured. Most sampling practices will “sample the sampleline,” which nonetheless does indicate abnormal releases from the core/plant surfaces.Redesigning existing reactor coolant sampling systems to provide adequate sampledelivery that considers flow kinetics (e.g., Sizewell B sample system) is not feasible formost domestic PWRs. At least one plant has adopted the practice of collecting samplesat 64

oC with steady flow through the existing sampling system as a compromise (17).

The main feature in obtaining reliable RCS corrosion product samples is to provide acontinually flowing system without disturbances and exposure to air/oxygen. SizewellB, for example, allows about two weeks for the sample line transients to disappear. Anacceptable alternative to an intermediate RCS sample is to use a CVCS sample collectedin an in-line filter holder, as used successfully at Vandellos II.


A-19

Flow changes cause transients in the standing soluble concentrations and hencechanges in the final equilibrium concentrations within the sample line, the latter beinginversely proportional to the sample line flow rate. UKAEA (18) attributes these effectsto changes in the interaction of soluble corrosion products with the oxides on thesample line surface as a result of the temperature profile along the sample line withchanges in flow rate. Additional studies carried out by UKAEA indicate thatdeposition of cobalt onto the sample line surfaces is mass transfer controlled at 300oCand 210oC, and kinetically controlled at lower temperatures (<120oC). Because of theseinteractions, what is measured may not be identical to the concentrations in the RCS inan absolute sense, but it should still be of significance in a relative sense.

Recognizing that the keys to obtaining reasonably accurate crud analyses in thecirculating reactor coolant are adequately purging the hot leg sample line and thelimitations on isolation valves, each plant should evaluate local sampling limitations todetermine how these might impact interpretation of sampling data. Purging a sampleline for only 30 minutes may provide reproducible results for trending purposes thatare nevertheless quantitatively biased.

5.0 Monitoring for Fuel Reliability

A committee of industry nuclear fuels experts developed a set of guidelines for utility-specific policies and procedures to secure improvements in fuel reliability. Theseguidelines are contained in Fuel Integrity Monitoring and Failure Evaluation Handbook(EPRI TR-108779, August 1998).

6.0 Primary-to-Secondary Leak Monitoring

Previous versions of these guidelines have addressed the importance of primary-to-secondary leak monitoring and calculation of leak rates as a portion of the primarywater chemistry program. In 1994, an industry committee was formed to develop a setof guidelines to serve as a standard program for managing primary-to-secondaryleakage. The guidelines were updated in 1997, and published as EPRI TR-104788-R1,PWR Primary-to-Secondary Leak Guidelines. Utility personnel can use the standardprogram as a template for developing their own site-specific programs. Industrypersonnel are referred to this document for additional details.


A-20

7.0 References

1. C. F. Baes and Mesmer, R.E., The Hydrolysis of Cations, John Wiley and Sons, NewYork, NY. 1976.

2. R. Murray and J. Cobble, "Chemical Equilibria in Aqueous Systems at HighTemperatures", IWC-80-25, Proceedings of the International Water Conference,Pittsburgh, PA. October, 1980 (Curve-fitted to information provided within).

3. P. Cohen, Water Coolant Technology of Power Reactors, American Nuclear Society,1980. (Curve-fitted to information provided within).

4. W. L. Marshall and E. U. Franck, "Ion Product of Water Substance, 0-1,000°C, 1-10,000 Bars. New International Formulation and Its Background", J. Phys Chem RefData, 10, 2, 1981 (Curve-fitted to information provided within).

5. Lange's Handbook of Chemistry, 13th Ed., John A. Dean, Editor, McGraw-Hill BookCo., New York, NY, 1985.

6. R.E. Mesmer, C.F. Baes, Jr., and F.H. Sweeton, Inorganic Chemistry, 11(3) 537, 1972.

7. PWR Primary Water Chemistry Guidelines, Revision 1, Electric Power ResearchInstitute, Palo Alto, Calif., Special Report, August 1988. NP 5960-SR.

8. Survey of Corrosion Product Generation, Transport, and Deposition in Light WaterNuclear Reactors. Electric Power Research Institute, Palo Alto, Calif. March 1979.NP-522.

9. K. Garbett and M. A. Mantell, “Elemental and Radionuclide Corrosion ProductMonitoring at Sizewell B PWR,” Presentation at PWR Primary Water ChemistryGuidelines Revision Meeting, Clearwater Beach, Florida. February 3 – 5, 1998.

10. A. Bates, Nuclear Electric, “Particulate Formation During RF02 at Sizewell B.”Presentation at PWR Primary Water Chemistry Guidelines Committee Meeting onShutdown Chemistry, Chicago, Illinois. May 19 - 20, 1998.

11. The Nature and Behavior of Particulates in PWR Primary Coolant. Palo Alto, Calif.:Electric Power Research Institute, December 1989. NP-6440.

12. M. Noe, et al., “Use of Ion Exchange Filters for Corrosion Product Sampling.”Proceedings: Workshop on Corrosion Product Sampling From Hot Water Systems.Electric Power Research Institute, Palo Alto, Calif., March 1984. NP-3402-SR.


A-21

13. D. A. Bridle, G. R. Brown, and P. A. V. Johnson, “The Application of TransitionMetal Ion Chromatography to the Determination of Elemental and RadiochemicalSpecies in PWR Primary Coolant.” Presentation at 1990 International Conference ofMeasuring Waterborne Trace Substances. Baltimore, Maryland: Electric PowerResearch Institute and National Institute of Standards and Technology. August 28 –30, 1990.

14. M. V. Polley and I. R. Brookes, PWR Primary Coolant Sample Lines – Problems withMeasurement of Corrosion Products and Experimental Proposals for Ringhals PWR.Gloucestershire, UK: Central Electricity Generating Board, July1987. GL13 9PB.

15. J. Roesmer, “Hot Reactor Coolant Sampling: Westinghouse Approach.”Proceedings: Workshop on Corrosion Product Sampling From Hot Water Systems.Palo Alto, Calif.: Electric Power Research Institute, March 1984. NP-3402-SR.

16. J. Roesmer, “An Assessment of the Composition of PWR Coolant Near theOperating Temperature.” Proceedings, JAIF International Conference on WaterChemistry in Nuclear Power Plants, Volume 2, April 19 – 22, 1988, pp. 482 – 467.

17. B. D. Fellers and D. Perkins, TU Electric, “RCS Sampling Experience, ComanchePeak Steam Electric Station.” Presentation at Robust Fuel Fuel/Water ChemistryWorking Group, Subgroup on Chemistry Diagnostics, Chicago, Illinois. May 20 - 21,1998.

18. K. Garbett of British Energy, comments on Draft Revision 4 of the PWR PrimaryWater Chemistry Guidelines, submitted October 26, 1998.

B-1

B CHEMISTRY CONTROL OF SUPPORTING SYSTEMS

1.0 Introduction

Since chemistry control practices in interfacing systems can impact on the chemistry ofthe reactor coolant, chemistry control in several such systems is discussed below.Specifically, this appendix discusses selected system designs, presents a rationale forchemistry control, discusses possible impacts on reactor coolant chemistry, and, whereappropriate, discusses industry experiences where impacts on RCS chemistry havebeen observed. Suggestions are provided on chemistry and radioactivity parameters tobe monitored and the frequency of monitoring.

2.0 Letdown Purification System

2.1 System Description

The letdown purification system is designed to control reactor coolant chemistry byremoval of impurities, corrosion products and radioactive species throughout the fuelcycle and during plant shutdowns. In this system, reactor coolant is passed throughfilters and demineralizers after temperature and pressure are reduced. The operatingparameters are typically 150 psig at ~120°F. Letdown purification components areconstructed of austenitic stainless steels.

All designs provide for remote replacement of filter cartridges and ion exchange resins.However, radiological considerations associated with handling and packaging of spentresins and filters may limit the operational lifetime of the filter or demineralizer.Typical designs include two or more cartridge filter assemblies, one in service and oneor more in standby, and two or more demineralizer vessels. Most plant designs includepost-demineralizer filters.

Purification demineralizers are designed to remove soluble ionic impurities and tofilter suspended material. Normally one mixed bed demineralizer with the cation resinin the lithium form and the anion resin in the borate form (1:1 equivalent mixture) isutilized to remove ionic impurities. In early cycle operation, when significant lithiumproduction occurs from the 10B(n,α)7Li reaction (approximately 0.14 ppm/day at 1000

Chemistry Control of Supporting Systems

B-2

ppm boron), and when routine RCS dilution cannot maintain the lithium concentrationwithin specification, another vessel (non-lithiated cation resin or mixed bed) isperiodically placed in service to reduce lithium concentrations to within operatinglimits. Near the end of the cycle, when routine control of RCS boron by dilution wouldproduce excessive liquid radwaste, deborating ion exchangers or a boron thermalregeneration system (BTRS) may be employed. The BTRS is also designed to operatethroughout the fuel cycle. Finally, during shutdown operations, a cation resin bed isoften placed in service to both delithiate the RCS and to remove the Co-58, Co-60 andnickel that are released. If significant fuel failures occur, hydrogen-form cation resinmust be used to remove Cs-134 and Cs-137 since lithiated cation resin has little affinityfor cesium in lithiated streams. This action also will remove lithium, and lithiumadditions may be needed to compensate.

2.2 Selection Criteria for Purification Filters and Ion Exchange Resins

Ion exchange resins and filters should be procured to meet utility and NSSS vendorrequirements. The quality of the resin should be specified and verified prior to use.The quality of the resins and filters should always be high due to the significant costsassociated with disposal.

The industry move to extended life cores and elevated coordinated pHT chemistryresults in reduced equilibrium fluoride and chloride capacity of the mixed beddemineralizer as a result of the higher concentration of hydroxide and ionized boronspecies in the coolant. Consequently, additional emphasis has developed onprocurement of nuclear grade, low chloride anion resin for application in the letdownsystem.

2.3 Performance Monitoring

Letdown demineralizer removal efficiencies for the following parameters are typicallymonitored:


B-3

Mixed Bed Demineralizer

Parameter Frequency(1) BasisChloride 1/W RCS impurity controlSulfate -- Resin ingress and RCS impurity controlIsotopics(2) 1/W Demineralizer performance

De-lithiating Ion Exchanger

Parameter Frequency(1) BasisLithium As required Lithium management

Deborating Ion Exchanger

Parameter Frequency(1) BasisBoron (3) Boron ManagementChloride (3) Impurity control

___________

(1) The frequency is normally increased significantly during reactor startup or shutdown.(2) Radioiodines, Cs-134, Cs-137, Co-60, Co-58, plus others of specific interest to the site.(3) At a reasonable frequency during application.

Demineralizer performance relative to chemistry and radioactivity parameters shouldbe reviewed and compared to normal performance to identify unusual trends. Ifunusual chemical or radioactivity concentrations are noted in the RCS, the performanceof the letdown ion exchangers should be assessed relative to the need for correctiveactions.

Other parameters monitored on the letdown demineralizers and filters are:

Parameter Frequency Basis

Differential Pressure * Monitor corrosion product loading

Radiation Levels * Minimize processing costs andexposure related problems.

____________

*As required.


B-4

Analytical results can provide input as to the source of differential pressure buildupand the isotopics responsible for the radiation fields.

The performance of the demineralizer is either monitored by percent removal ordecontamination factor (DF). The percent removal is defined as:

%Removal =CIn −COut

CInx 100

DF is defined as:

DFCCIn

Out=

Either parameter is acceptable for trending performance. A DF of 1 corresponds to 0%removal; a DF of 100 corresponds to 99% removal. At steady state conditions, a mixedbed demineralizer in the lithium/boron form will have a DF of 1 for these species.

It is important to note that the DFs for radioactive and non-radioactive forms for thesame element may not be the same. For example, the DF for sodium may be 1 (e.g. 0%removal) while the DF for Na-24 may be 500 (e.g. 99.8% removal). The reason for thisdifference is the time necessary for an ion to transit the bed. A short-lived nuclide maydecay to insignificance in the time taken. Transit time is a function of the selectivity ofthe resin for the ion. The more highly retained the ion, the longer it takes for the ion totransit the bed. The effect of transit time and half-life frequently can be seen in the DFvariation for the iodine isotopes.

Because of parent-daughter relationships, some radionuclides entering the bed can beretained on the bed and subsequently decay to a product that is not retained within thebed. For example, I-133 and I-135 decay to the daughter products Xe-133 and Xe-135,respectively. Normally this is not a concern since the xenon daughter products areentrained in the effluent and returned to the RCS. If a bed highly loaded withradioiodine is isolated, the ion exchange vessel will build up a considerable inventoryof the noble gas decay products. Another example of parent-daughter relationships isRb-88 which may appear to have an unusually low DF as a result of the parent, Kr-88,in the effluent sample decaying to the daughter, Rb-88, in the sample container prior tosample counting.

The inventory of activity on the bed can be estimated from the expression:

Ai =Ci • F•∈i

λi(1− e−λit )


B-5

where:

Ai = µCi (or Bq) of activity of isotope i on the bed,Ci = activity concentration of the isotope at the bed inlet, µCi/mL (or

Bq/mL)F = letdown flow to the demineralizer, mL/hr∈i = efficiency of removal of isotope Iλi = decay constant of the radionuclide, h-1, andt = operating time, hours

2.4 Industry Experience

Experience 1. Plants have experienced intrusions of resin from purificationdemineralizers during plant operation when the letdown filter is bypassed. In onecase, the cause of the event was suspected to be an inadvertent backflush of the cationexchanger during a valve change. Another resulted from a damaged ion exchangerretention element.

Experience 2. Several plants have experienced chloride elution problems from ionexchange resin when boric acid concentrations were increased at refueling. Theresultant chloride concentration exceeded Technical Specification limits. The elutionwas related to the use of anion resin with significant levels of residual chlorides. Lowleachable chloride resins and tighter adherence to purchase specifications haveminimized events of this type in recent years. Impurity elution can also occur if a bedhas equilibrated at EOC lithium/boron concentrations, and an attempt is made to reusethe bed at the beginning of the next cycle.

Experience 3. Plant testing has revealed significant leachable impurities (40-200 µg/gof material) from charging pump packing material. This should be considered inevaluation of impurity sources when investigating anomalous RCS chemistry results.

Experience 4. Unintended gradual boron dilutions have occurred at full power at someplants as a result of incomplete flushing of purification demineralizers, and placinginadequately saturated demineralizers in service. For example, one plant exceeded itslicensed power level for 90 minutes due to an incomplete flush of the letdown headerfollowing a boric acid soak and rinse of a new mixed bed ion exchanger. A section ofthe header remained filled with low boron concentration reactor coolant which wasadded to the reactor coolant.

Another plant operating at 100 percent power experienced a power increase to 101.1percent after a cation ion exchanger was placed in service. Water in the ion exchangervessel and piping contained less than 100 ppm boron. When the ion exchanger was


B-6

placed in service the low boron concentration water was added to the reactor coolantsystem.

Experience 5. During startup of a plant outside the U.S., operations transferred 2,500liters of lithium hydroxide solution into the reactor coolant. Nine hundred liters hadbeen projected as the amount required. With the large injection, reactor powerincreased immediately from 8.5 percent to 12 percent.

3.0 Volume Control Tank


The volume control tank (VCT) and the charging system are designed to maintain RCSsystem volume during periods of minor system leakage and diversion and to supportthe RCS chemistry control program. The system also provides a mechanism for controlof water radiolysis during operation and for removing hydrogen and fission gasesduring shutdown evolutions. Oxygen reduction is performed by introduction ofhydrogen into the coolant via the VCT to suppress the radiolytic production of oxygen.As the letdown liquid from the purification system is sprayed into the VCT, itequilibrates with hydrogen and other gases in the tank vapor space. A specifiedhydrogen pressure is maintained in the VCT to establish the desired hydrogenconcentration in the RCS.

The VCT is a stainless steel tank that is continuously filled with letdown from thepurification system. During operation, the vapor space of the VCT is filled withhydrogen and maintained at 15 to 30 psig. Generally, the VCT vapor, which may alsocontain fission gases, is periodically purged to the waste gas system.

Just as hydrogen is added to the RCS via solubilization from the VCT vapor, the systemcan be degassed of hydrogen by reversal of the process. Prior to or subsequent toshutdown, the VCT vapor is purged with nitrogen and letdown flow increased. As thesystem flow continues and the vapor space is maintained free or near-free of hydrogen,RCS hydrogen concentrations decrease in accordance with Henry's Law (discussedbelow).

Hydrogen removal practices are discussed in detail in Volume 2.


B-7

3.2 Chemistry Control and Technical Basis

The VCT is typically monitored for the following parameters:

Parameter BasisComposition Estimation of RCS H2 and knowledge of other

gases presentPressure Estimation of RCS H2

Radiogas Knowledge

The solubility of a gas in a liquid can be expressed using Henry's Law which states thatthe solubility is proportional to the partial pressure of a gas above the liquid.Mathematically, this is expressed by:

XPHii

i=

where: Xi = mole fraction of the gas i in the liquidPi = partial pressure of gas i in the gas phase (atmospheres)Hi = Henry's Law constant (atm/mole fraction)

The Henry's Law constant is a function of temperature. Values of the Henry's Lawconstant for several gases of interest over the temperature range of interest in the VCTare given in Table B-1. The constant has been converted to more useful units for ease ofapplication.

Table B-1Solubility In Water For Various Gases

Presented as the Reciprocal of Henry’s Law Constant (1/H),cc (STP)1/kg-psia

Gas 100°F 110°F 120°F 130°F 140°F 150°FHydrogen 1.14 1.12 1.12 1.12 1.13 1.14Nitrogen 0.830 0.78 0.756 0.730 0.717 0.700Oxygen 1.62 1.52 1.44 1.38 1.32 1.28Krypton 2.92 2.70 2.51 2.36 2.23 2.14Xenon 4.78 4.32 3.94 3.65 3.40 3.22Helium 0.568 0.572 0.580 0.592 0.609 0.627

_______________________

1. STP - 273.15°K and 1.0 atmosphere.


B-8

There are limited plant data on the kinetics of gas equilibration via the spray into theVCT. However, the kinetics are generally sufficiently rapid that operators have littledifficulty maintaining adequate RCS hydrogen inventory with minimal chemistrysurveillance of VCT vapor.

Often it is found that Henry's Law applied to the VCT predicts a concentration in theRCS that is higher than actually measured in a hot leg sample. There are severalreasons that this occurs. The principal reason is the loss of hydrogen to the secondaryside via diffusion through the steam generator tubes. Other factors include loss ofhydrogen by boron dilution, low level oxygen in the makeup water that is consumedby hydrogen in the radiation field, lack of equilibration in the VCT and in some casespressurizer vent leakage. The differences between the Henry's Law prediction and themeasured hot leg value can be on the order of 5 cc(STP)/kg. The VCT gas space alsotypically contains significant concentrations of long-lived noble gases. For example, aVCT at equilibrium will have a vapor-to-liquid concentration ratio of Xe-133 ofapproximately 12 to 1 at 100°F.


Experience 1. On plants with oxygenated makeup water, the VCT vapor cantheoretically increase in oxygen concentration to approximately 0.5% and a four timeshigher volume percent of nitrogen, assuming a makeup rate of 2% of the letdown flowrate. This can lead to continuous, low-level oxygen contamination of the water chargedinto the RCS. Such a scenario can also lead to oxygen contamination of the waste gasholdup tanks which are limited to <2% O2.

Experience 2. For plants with de-oxygenated makeup water utilizing a nitrogen cap onthe makeup water storage vessel, N2 can build up in the VCT. One utility purges theVCT approximately biweekly to maintain the hydrogen volume fraction greater than95% and reduce the radiolytic generation of ammonia in the RCS. When ammonia ispresent, it must be factored into the RCS conductivity/pH balance.

4.0 Pressurizer


The purpose of the pressurizer is to control RCS pressure. In addition, the pressurizersystem maintains the proper primary coolant inventory during steady state operationand allows for expected transients. The lower portion of the pressurizer vesselincorporates electrical heaters and the upper or gas portion of the vessel is equippedwith a water spray nozzle. RCS pressure is maintained by affecting the water/steamequilibrium with a combination of electric immersion heaters and a fine spray of


B-9

relatively cool water. Changes in coolant volume are handled by allowing the waterlevel in the pressurizer vessel to vary directly with RCS Tavg. Adjustments to the liquidlevel in the pressurizer vessel are accomplished by varying the RCS system chargingand letdown flow.

The pressurizer is connected to the hot leg loop by a surge line. During reactoroperation, with the system pressure at ~2235 psig, the hot and cold leg temperaturesare approximately 600°F (316°C) and 550°F (288°C), respectively. The primary coolantwater and pressurizing gas within the pressurizer are maintained at an equilibriumsaturation temperature of ~650°F (343°C). The cooler water in the cold leg is used forthe pressurizer spray when necessary to control pressure increases. A continuous sprayof ~1 gpm is provided to prevent excess cooling of the spray system, to limit the effectsof thermal shock to the lines and to aid in the equalization of water chemistry betweenthe pressurizer and the remainder of the RCS. The maximum spray rate is ~800 gpm.The pressurizer system is primarily constructed of austenitic stainless steel.


During some operating evolutions, reactor coolant is circulated through the pressurizer,and RCS and pressurizer liquid samples are simultaneously analyzed to ensure boronequilibrium. This is not the case in normal plant operation, and routine surveillance ofthe pressurizer chemistry is generally not necessary. However, it is the general practiceat many PWRs to monitor the boron and gas composition.

Provisions for sampling the gas or liquid space are often provided in the NSSSsampling system. Venting of the pressurizer gas space through the NSSS samplingsystem is possible, but ultimate control of pressurizer water and steam chemistrygenerally is only possible through RCS bulk water control.

During startup, when the pressurizer heaters are first used, the resultant boiling stripsammonia, nitrogen, oxygen and other gases from the liquid and they accumulate in thevapor space (note that pressurizer oxygen concentration must be controlled to the samevalues as the rest of the RCS during startup). During normal power operation, thepressurizer may be a source of impurities since these gases gradually re-enter thepressurizer liquid as predicted by Henry's Law. It should be noted that the pressurizervapor also accumulates hydrogen gas. This is particularly important since, if thepressurizer vent leaks, hydrogen will be stripped from the RCS, with the result that thesteady-state hydrogen concentration in the coolant could become significantly less thanpredicted by Henry's Law applied to the VCT considering known hydrogen losspathways. During shutdown, hydrogen return to the RCS may occur from thepressurizer (see Volume 2).


B-10


Experience 1. Stress corrosion cracking of Alloy 600 instrument nozzles and heatersleeves has been observed, and has been attributed to high stress, high cold work, andhigh-temperature. It has also been speculated that oxygenated water resulting frominadequate venting or chemical treatment during heatup may have been a factor.Oxygen should not be a problem if properly managed by chemical treatment duringheatup.

Experience 2. Excessive pressurizer leakage can significantly affect the ability tomaintain hydrogen inventory. Plant parameters that indicate pressurizer leakage (e.g.,containment air monitors) should be closely monitored when RCS hydrogen data areunusual.

Experience 3. Use of a cleaner containing 1,1,2-trichloroethane on emergency injectionpumps at one PWR led to ppm-level chloride contamination in shutdown cooling waterand in the pressurizer. At elevated temperatures, this compound will decompose toyield chloride.

5.0 Boric Acid Storage


The boric acid storage system contains solutions of soluble reactor poison that arecontinuously available to the RCS, refueling water storage tank (RWST) and spent fuelpool (SFP). During initial operation, the desired concentration is achieved bydissolving nuclear grade boric acid in demineralized water. Subsequent makeup maybe provided by newly prepared batches of boric acid and/or by recycling of RCSletdown that has been diverted to storage tanks. The diverted RCS liquid is typicallypassed through ion exchangers to remove impurities, stored in large waste storagetanks, and fed to evaporators where the liquid is concentrated until the boric acidconcentration meets specific plant requirements. Some plant designs also include ionexchangers to further purify the concentrated boric acid solution prior to transfer to theboric acid storage tanks. Many plant operators have elected to dispose of the boric acidto minimize problems associated with the concentration process.

The basic components of the boric acid storage system are: the boric acid storage tanks(BASTs), the recycle holdup tanks, evaporator feed ion exchangers and one or morerecycle evaporators. Most components are constructed of austenitic stainless steel.

Some plant designs and operating practices require recovered boric acid to beconcentrated to as high as 13%. Such high concentrations require heat tracing thesystem to prevent precipitation. Continuous recirculation is often required to prevent


B-11

crystallization in areas of local cooling. Operational problems can be encounteredwhenever heat tracing of any part of the system is lost, since the highly concentratedsolution will crystallize whenever the temperature drops below the saturationtemperature. Resolubilization upon temperature recovery is very slow. Plantsdesigned to operate at a 4% boric acid concentration are not subject to such problems.


During the recycling of RCS letdown, any impurities and activity not removed by theion exchangers will concentrate in the evaporator bottoms and will be transferred to theboric acid storage tanks. If the plant is not equipped with boric acid ion exchangers,the contaminants will be subsequently returned to the RCS, RWST or SFP. Inevaluating the recycle of concentrated impurities, the utility must ensure impurityspecifications for the concentrate reflect normal levels after dilution in the RCS.

Boric acid storage tanks have three sources of chemical contaminants: (1) makeupwater, (2) virgin boric acid additions, and (3) evaporator bottoms. Each of thesesources has potential for contamination which may eventually enter the RCS throughnormal boration steps. It is important to recognize this potential and to develop asystematic analytical program that minimizes the likelihood of RCS contamination aswell as protects the components of the boric acid system materials. At the firstindication of unusual chemistry in the stored boric acid, the feed sources should beinvestigated for contributing parameters.

Makeup water and nuclear grade boric acid may have trace quantities of aluminum,calcium, magnesium, and silica. Many oxides and silicates of aluminum, calcium, andmagnesium have negative temperature coefficients of solubility, e.g., they will depositpreferentially at the hottest locations. In the borated water storage system, they mayform deposits on the immersion heaters and impair their ability to transfer heat into thesystem. Also, if these species are transferred into the RCS, they may deposit on the fuelrods and adversely impact fuel performance.


B-12

Chemistry parameters that are suggested for monitoring include:

Parameter Frequency(1) Limit, ppb(2) BasisBoron 1/W -- Technical specificationChloride 1/W 150 Impurity controlFluoride 1/W 150 Impurity controlSulfate 1/W 150 Impurity controlSilica 1/M 5000 Impurity controlCa, Mg, Al 1/M 200 Impurity controlIron As Required -- CorrosionIsotopics 1/M -- Knowledge

1. Routine monitoring is recommended for plants that recycle boric acid. Application of strictpurchase specifications and periodic surveillance may be acceptable where boric acid is notrecycled.

2. Limit for 4% boric acid. Limit can be increased proportionally for higher percent boric acid.

Boron is a Technical Specification item. Chloride and fluoride specifications relate tocontrol of these impurities in the RCS which are currently RCS Technical Specificationitems.

Sulfate falls into the category of a chemical species known to increase the tendency forIGSCC of stainless steel and Alloy 600. Sulfate can enter the system via the BAST.However, a more likely source is ingress of letdown demineralizer resin.

Silica controlled to within guideline limits is generally not a problem unless boiling isoccurring in the core, and hardness elements (i.e., calcium, aluminum and magnesium)are present. Under these conditions, deposition on fuel surfaces may occur. Suchdeposits can act as a heat transfer barrier and/or provide a deposit where lithium andboron can concentrate. This can result in an adverse impact on fuel performance andfuel cladding integrity. In addition, there has been some concern that hardnesselements can form deposits on the immersion heaters in the borated water storagesystem.

Many plants do not recycle boric acid. Some plants that recycle boric acid haveexperienced elevated impurity concentrations in evaporator concentrates. This occursmost often at the end of the fuel cycle when the boron concentration in the RCS bleed isrelatively low and the required concentrating factor very large. Reasons for notrecycling boric acid include items such as high maintenance on evaporators, B-10depletion, silica return to the RCS for plants utilizing Boraflex fuel racks and lack ofrestrictions on boron release to the environment.


B-13

Excessive iron in the BAST can result in a turbid solution upon dilution, such as theflooding of the refueling cavity. It may also be an indication of excessive corrosion ofmaterials in the boron recycling and storage system.

Organic materials containing chlorine, fluorine or sulfur have the potential to breakdown into detrimental ionic species. Post UV ion chromatography is a demonstratedmethod for quantifying concentrations of organically bound chloride, sulfur, fluorideand nitrogen.


Experience 1. Recycled boric acid has led to elevated concentrations of impurities inthe boric acid storage tanks. If the parameter that is elevated is not included in thechemistry surveillance program, high concentrations could occur in the RCS and theSFP. This has been a significant problem at plants that minimize waste releases andhave elevated silica from Boraflex. In some cases silica concentrations in the boric acidstorage tanks have risen to as much as 20 ppm.

Experience 2. Loss of heat tracing in the boric acid recycle system and eventualcooldown of the concentrated solution has lead to boric acid crystallization. This hasplugged feed lines and disabled the system.

Experience 3. An increase in RCS chloride concentrations at one plant began aninvestigation that discovered about an inch of oil on the surface of the BASTs. Thecause was found to be a cracked diaphragm in the recirculation pumps. One of theevent precursors was an observation by plant operators that the pump requiredexcessive oil fills without normal symptoms of oil leakage.

Experience 4. At two plants the recycled boric acid had increased levels of silica in theboric acid storage tanks and subsequently in the RWST and RCS. Boraflex material inthe spent fuel racks was identified as the source of the silica. This resulted in asituation requiring the utility to evaluate silica impact versus increased radwaste curiesand boron release in plant effluents.

6.0 Refueling Water Storage Tank


The refueling water storage tank (RWST) provides a source of borated water to therefueling pool, spent fuel pool, safety injection tanks and the RCS during accidentconditions. It is also used for refilling the RCS during startups at some plants. Theborated water is maintained at the concentration required by Technical Specifications to


B-14

ensure that the reactor will remain subcritical during core alterations and accidentconditions.

The RWST must provide a large volume of borated water to the operating plant.Recirculation pumps are provided at some sites to ensure a homogeneous boronsolution. Makeup to the RWST is usually provided from the boric acid storage tanksand primary makeup water supply. If necessary, cleanup or demineralization normallycan be provided by the spent fuel pool (or refueling water purification system)demineralizer. As a result of the large volume of the tank (~500,000 gal) and relativelylow cleanup flow (typically 100 gpm), the cleanup half-life is very long (~2.4 days) evenat maximum flow.


RWST chemistry parameters that are suggested for routine monitoring are listed below.Prior to use of RWST water for RCS refill it is suggested that the parameters listed inthe table in Section 5.2 be checked.

Parameter* Frequency** BasisBoron Tech. Spec. Technical SpecificationChloride 1/M Impurity source identificationFluoride 1/M Impurity source identificationSilica 1/M Potential water turbidity concernTurbidity or TSS 1/M Water clarityIsotopics 1/M Knowledge

* Monitor for sodium if there are potential sources of sodium that could contaminatethe RWST water.

* * Increase frequency of surveillance of RWST prior to use for fill of the RCS.

The primary mechanism for controlling the water quality of the RWST is to ensure thatthe quality of makeup water and concentrated boric acid solutions are acceptable priorto addition to the tank. It should be noted that optical clarity of refueling water is animportant aspect of the refueling operations.

Boron is monitored as a Technical Specification item. Chloride and fluoride aremonitored to provide a basis for impurity source identification. Since this water willenter the RCS during refueling, and may be used for RCS refilling, it is required thatthe maximum concentration of fluoride and chloride not exceed the RCS operatingTechnical Specification limit for these species.


B-15

Silica is monitored to ensure the reactor coolant system silica specification can be metfollowing refueling operations. High concentrations of colloidal silica can alsocontribute to reduced water clarity. Concentrations of silica should be limited asneeded to support return to power operation following refueling. Turbidity or totalsuspended solids measurements provide an indication of the optical clarity of thewater. Species contributing to turbidity include colloidal silica, corrosion products, orother suspended matter. Turbidity in excess of 0.05 NTU is cause for sourceinvestigation.

Other parameters monitored in the industry include sodium (spray additive systemleakage), pH, lithium, ammonia, sulfate, hardness elements (Ca, Mg and Al), TOC,specific conductivity and tritium.

Due to the time required for effective cleanup of the RWST, careful planning andscheduling is needed to prepare for a refueling outage. Approximately a month priorto a planned refueling outage, expansion of the normal chemistry monitoring programshould be considered to allow sufficient time for cleanup. If the SFP and RWST share acleanup system, this factor needs to be considered for cleanup scheduling.


Experience 1. Lack of optical clarity in the reactor cavity water, resulting fromaccumulation of finely dispersed solids in the RWST or from corrosion product releaseassociated with fuel movement significantly impacted refueling operations.

Experience 2. After the refueling cavity was filled in preparation for refueling, ionexchange resin was found in the reactor vessel and reactor cavity. The resin hadapparently collected in either the refueling water storage tank or the emergency corecooling system suction header and was flushed to the reactor system during cavityfiling. Twenty days of cleanup were required to remove the resin.

Experience 3. High concentrations of silica in the RWST have been successfullyremoved using reverse osmosis technology with minimal impact on boronconcentration.

7.0 Spent Fuel Pool Cooling and Cleanup System


The spent fuel pool cooling and cleanup system is designed to provide the means ofsafely cooling and storing spent fuel assemblies until removal to an on-site or off-site


B-16

storage facility. The spent fuel storage pool and fuel transfer canal are constructed ofreinforced concrete and are lined with stainless steel or coated carbon steel.

A spent fuel storage pool can hold 300,000 to 600,000 gallons of water, which providessufficient depth so that the surface dose will be ~2.5 mrem/hr when moving a fuelassembly over the storage rack.

The system's demineralizers and filters are designed to provide adequate purificationto permit unrestricted access of plant personnel to the spent fuel storage area and tomaintain optical clarity of the spent fuel pool water. Optical clarity is maintained byuse of the system's skimmers, strainer and skimmer filter. At some sites, portablefiltering systems are used in the reactor cavity during refueling operations.

Normal makeup for the spent fuel pool comes from the RWSTs, with alternate makeuppaths from the Boron Recycle System holdup tanks via the transfer canal, or from thedemineralized water system. An emergency source of makeup can come from thecondensate storage tanks or the service water system.

Several types of spent fuel racks are utilized in the industry. One type of poison spentfuel storage rack is composed of a welded array of stainless steel tubes bearing Boralpoison material. Boral consists of boron carbide (B4C) particles embedded in a matrixof pure aluminum formed into a plate and clad with aluminum on both sides. Asecond type of fuel rack common in the industry contain Boraflex. Boraflex consists ofboron carbide dispersed in a silicone polymer. In recent years, elevated silica has beenobserved in fuel pools utilizing such racks as a result of silica solubilization. Plantswith this problem should minimize mixing of SFP and reactor cavity water. The onlypractical way of reducing silica concentration in the presence of boric acid is reverseosmosis, which is very expensive.

7.2 Chemistry Controls and Technical Basis

Chemistry controls are required to maintain water clarity, minimize dissolved andairborne activity and minimize corrosion of stored fuel assemblies and of constructionmaterials. Water clarity should be such that the lettering on the fuel bundles can beread during fuel movement. In addition, consideration must be given to the fact thatfuel pool water mixes to some extent with the reactor water during refueling operation.Significant transfers of liquid can occur if the levels in the two pools initially are notequal.


B-17

Chemistry parameters that are suggested for monitoring include:

Analysis Frequency Limit BasisBoron, ppm Tech. Spec. -- Technical specification,

knowledgeChloride andfluoride, ppb

1/M <150(inlet)

Impurity control anddemineralizer performance

Sulfate, ppb 1/M -- Impurity controlSilica 1/M -- Status of Boraflex fuel storage

racks, and impurity controlGammaIsotopics

1/W -- Fuel cladding status anddemineralizer performance

Turbidity NTU(or TSS)

1/M -- Water clarity trending

Boron is a Technical Specification item at some PWRs. The boron specification is set tomaintain subcriticality and ensure safe storage of the fuel. At most PWRs, there are nochemistry related Technical Specification limits for the spent fuel pool. However, sincethe fuel pool water is transferred to and from the RWST which does have a TechnicalSpecification boron lower limit, the same limit has often been applied to the fuel poolcoolant. For similar reasons limits have been specified for chloride and fluoride. Also,if dry fuel storage is practiced, cask loading in the SFP may require a minimum boronconcentration per Technical Specification.

Chloride, fluoride and sulfate are monitored to reduce the potential for stress corrosioncracking of the fuel pool and related system materials and to reduce the potential forcontamination in the reactor coolant system during refueling operations. Ifconcentrations increase above normal, the RWST and/or discharge of the fuel pooldemineralizers should be analyzed. Close attention to material controls in the SFP areacan be used to prevent or identify an impurity ingress pathway.

Turbidity (or TSS) is monitored and controlled to maintain adequate visual clarity formovement of fuel assemblies and to minimize the potential for contamination of thereactor coolant system during refueling operations. Excessive turbidity is cause forsource investigation. The general radionuclide gamma scan is performed to monitorfuel integrity and ensure that radiation levels in the fuel pool area are kept as low asreasonably achievable.

Other analyses performed by utilities include pH, hardness elements (Ca, Mg and Al),specific conductivity, lithium, TOC and tritium.


B-18


Experience 1. IGSCC failure of top nozzles of fuel elements stored in spent fuel poolwas related to the release of sulfur compounds (as sulfate) from the cleanup resin bedas a result of exposure to hydrogen peroxide in the spent fuel pool. The release wasfrom cation resin located downstream from the mixed bed resin. A historical review offuel pool chemistry showed there had been a decreasing pH and increasingconductivity. The spent fuel pool resins were tested for leachables and the cation resinshowed high conductivity and sulfates with low pH (1).

Experience 2. A plant with Boraflex spent fuel storage racks has experienced highreactive and colloidal silica in the SFP. Above 14 ppm reactive silica, the turbidityincreased noticeably. Also, TSS (Total Suspended Solids) samples of the pool watershowed that 0.3 NTU turbidity correlated to about 2 ppm TSS. The colloidal silicawhich was thought to be responsible for the reduced water clarity could be removedwith sub-micron filtration. The current policy of the plant is to only clean up andmaintain the suspended silica when clarity is needed.

8.0 References

1. OE 5942, Top Nozzles Found Detached from Fuel Assemblies Located in the SpentFuel Pool, OER Number OTH 93-082, LCTS Number: 9301265-936, dated 22 April1993.

C-1

C STATUS OF ENRICHED BORIC ACID (EBA)

APPLICATION

1.0 Introduction

Natural boric acid (NBA) is used in primary coolant of a PWR as a soluble reactivitycontrol agent. Dissolved boric acid is referred to as a soluble poison or chemical shimbecause the B-10 isotope has a high capacity for absorbing thermal neutrons. However,natural boron contains only 20 atom percent of the B-10 isotope. The remaining 80% isthe B-11 isotope, which has only a small cross section for absorption of thermalneutrons. Enriching the content of the B-10 isotope would reduce the boric acidconcentration in operating PWR plants to only a fraction of that presently used andwould provide other benefits. However, enriched boric acid (EBA) is costly, which hasimpeded introduction of this technology.

2.0 Summary of EPRI Studies

The first EPRI assessment of the costs/benefits of EBA was performed in 1989.Potential benefits were found to be: [1] eliminating the need for line heat tracing; [2]allowing increases in reactor coolant pH (which decreases crud deposition on the coreand radiation level buildup); [3] allowing longer fuel cycles; and [4] reducingmaintenance of components that can be affected by boric acid recrystallization. Thelargest benefit found was that associated with a reduction in personnel exposure.These benefits must be weighed against significant conversion costs, the high costs ofEBA, the operational time required to change a large amount of natural borated waterinto enriched boron liquids, the potential reconditioning of existing equipment for boricacid recovery, and the cost and complexity of purchasing and maintaining newequipment.

The potential for EBA was recently reassessed because PWR utilities are seeking toextract more energy from nuclear fuels by extending the fuel cycle length and operatingat higher power. Longer fuel cycles require higher lithium concentrations (e.g., up to3.5 ppm) for longer periods of time to maintain the optimum pH in the coolant. Thesechanges could accelerate fuel cladding corrosion. Also, operation at higher power has

Status of Enriched Boric Acid (EBA) Application

C-2

led to nucleate boiling, a necessary factor for the occurrence of the axial offset anomaly(AOA), which has led to plant deratings. EBA permits an increase in the pH level inthe primary coolant, and a decrease in the lithium concentration, thereby amelioratingsome of the increased demands on the fuel. The new study evaluated the effects andthe risks of switching to EBA on nuclear fuels, structural alloys, fluid systems, andnuclear safety. The impact of switching to EBA on equipment, plant and fuel systemdesigns, and operating procedures also was determined. This information was used toidentify plants with the desired design features for implementing an EBAdemonstration. A radiation transport and deposition code was used to estimate theoccupational radiation exposure savings that would be realized by switching to EBA.

The majority of calculated savings associated with EBA are due to a reduction inpersonnel exposure. While this advantage does not offer sufficient justification forimplementing EBA technology, it appears that significant benefits can be realized fromthe improvement in fuel performance, although such benefits are more difficult toquantify. The use of EBA results in a lower absolute boric acid concentration andallows a higher pH to be maintained over a larger portion of the fuel cycle. Thesechanges are expected to reduce crud deposition and the precipitation of lithiummetaborate. EBA is thus expected to decrease the probability of fuel failures byreducing the crud loading, and a lower crud loading likely ameliorates AOA.However, one possible concern regarding the use of EBA has not been fully resolved.If lithium metaborate should precipitate despite the lower lithium concentrationsassociated with the use of EBA, the higher B-10 content of the precipitate mayaggravate AOA. Ongoing studies are evaluating the potential risks and benefits beforea field trial is implemented at a U.S. plant.

3.0 Plant Demonstrations of EBA

Four Siemens designed PWRs are now using EBA at 30% enrichment to address issuesrelated to operation with mixed-oxide fuel. No US PWR has used EBA to address theissues noted above. EPRI will address plant-specific issues at utility sites that mightconsider hosting a plant demonstration of the technology.

4.0 References

1. Enriched Boric Acid for PWR Application. Electric Power Research Institute, PaloAlto, CA. September, 1989. NP-6458.

2. Re-Evaluation of the Benefits of Implementing Enriched Boric Acid. Electric PowerResearch Institute, Palo Alto, CA. March, 1998. TR-109992.

D-1

D CVCS FILTER AND RESIN OPTIMIZATION PRACTICES

1.0 Introduction

Few quantitative studies have been performed to optimize filtration and resin practicesin the CVCS. Different configurations limit generic recommendation. EPRI performeda limited plant survey in June 1995 to determine best practices. Completeness ofresponses varied since information was not always readily available or benefits from aparticular practice were difficult to quantify. Lack of a particular problem could beinterpreted as successful practices, when in reality cause and effects had not beenproperly established. However, the survey yielded a variety of good practices whichare summarized below.

2.0 Filtration Practices

Most PWRs have filters downstream of the CVCS ion exchangers, and some plants havefilters upstream of the ion exchangers and separate reactor coolant pump seal and sealwater return filters. Cellulose/paper, cotton, glass fiber, nylon, polypropylene,polysulphone, ceramic material, and stainless steel media are used, with 73% of theresponses from domestic PWRs in the EPRI survey reporting glass fiber post filters(negative effects on reactor coolant silica levels were not reported). The only negativeeffects reported for the filter media were dissolution of a ceramic material and poorperformance of paper filters in two separate responses.

A growing practice is to use submicron filters to retain crud particles with the expectedbenefits of lower plant dose rates, fewer discrete radioactive particle (DRP) migrationproblems, improved reactor coolant pump seal performance, and reduced effluentproblems for discharged liquid radwaste. As a minimum, post ion exchanger filtersretain resin fines with 25 micron media, whereas submicron filters (e.g., 0.2 micron)retain crud particles. Radiation levels on the housing of post-filters with submicronmedia verify that resin does not completely retain crud particles. The following mustbe considered when evaluating submicron filters:

• All costs (e.g., more frequent filter changeouts, dose and disposal costs) must bebalanced against improved component performance and possibly lower plant doserates.

CVCS Filter and Resin Optimization Practices

D-2

• High radiation levels can result in problems in changing the filters in certain specificplant situations.

• The accumulated crud is a potential source term for species which can be dissolvedfrom the filter and be re-released to the plant.

Use of “fine” filters, such as the CVCS filter, has limited theoretical value, since thecorrosion products removed will gradually dissolve in the cool letdown stream. Inaddition, letdown flow rates yield an RCS particulate removal rate several times lowerthan observed deposition rates. Particulate removal may prevent fouling of the CVCSresin or a buildup of system pressure drop, but would not be expected to provide long-term radiation field benefit. The case may be substantially different for RCP sealinjection filters, where removal of fine corrosion products may prevent damage beforedissolution would have removed them. Reference (1) provides an example evaluationof various strategies.

Despite this theoretical information, 38% of the domestic PWRs responding to thesurvey reported dose benefits and 50% of the domestic PWRs with seal injectionsubmicron filters reported improved reactor coolant pump seal performance. Asexpected, most of the positive benefits associated with using submicron filters wasdifficult to quantify. However, the following benefits of using submicron filters,(particularly in the seal water injection system) were reported:

• Fewer hot spots were noted after submicron filters were used.

• Reduced seal replacement resulted in a $35,000 per year overall savings, which didnot include potential increased plant availability.

• $20,000 to $50,000 per year in dose savings.

• Reduced seal injection filter changeouts.

• $211,000 per year savings with ultrafiltration over back-flushable prefilter and areduction from 26 to 5 cavity filters during the previous outage.

Effects on CVCS resin performance, DRP migration, and reactor coolant crud activitylevels were not quantifiable from the responses.

The survey indicated two potential concerns with accumulated crud activity on filters.One plant reported that a 5-micron filter was not changed during a refueling outage,and the letdown piping dose rate increased from 130 to 300 mR/hr and the reactorcoolant Co-58 level increased in three months; conditions returned to normal within sixweeks after the expended filter was replaced with a 0.2 micron filter. Another PWRreported changes in the RCP seal water injection filter differential pressure (increase


D-3

followed by a decrease) after a lithium hydroxide addition to the plant, with anincrease in seal water return flow rate. This was later determined to be the pressureincrease resulting from filtration of LiOH granules which were undissolved during theaddition. These two experiences indicate that material accumulated on filters is apotential source term and that periodic filter changes are necessary to minimize re-release, and chemical additions may impact crud movement or dissolution.

In conclusion, using submicron filtration in both pre- and post-filters in the CVCS mayprovide short term benefits of crud removal with subsequent removal of the dissolvedcorrosion products by ion exchange. Experience suggests that filter pressure drops anddose rates must be monitored at a suitable frequency to indicate changeoutrequirements, and that use of submicron filters as RCP seal injection filters may havequantifiable benefits.

3.0 Resin/Ion Exchange Practices

All PWRs have at least two ion exchangers in the CVCS (or equivalent system) forchemical impurity control and for radionuclide removal. Some plants have several ionexchangers which can be operated in parallel. Plants with Westinghouse designsusually have a cation bed for lithium, cesium, and sodium removal which can beoperated in series with a bed used for normal purification. A common design also usesseparate beds downstream of the normal purification train which can be valved in toremove boron near the end of a fuel cycle. The two (or sometimes more) beds used fornormal purification contain cation resin in the hydrogen or Li-7 form, and anion resinin the hydroxide form which must be converted to the borate form prior to use. Sincereactor coolant pH is maintained by Li7OH, at least one of the beds used for normalpurification must contain cation resin in the Li-7 form. The resin in the cation vessel isinitially in the hydrogen form, and the resin in the deborating beds is in the hydroxideform. Since Li-7 is produced from B-10 by neutron absorption, coolant lithium levelsmust be controlled by removing the lithium. Ion exchange in addition to feed-and-bleed is used to maintain coolant lithium levels within the desired band. Thedifferences in designs and different operating philosophies have resulted in severalpractices in operating the ion exchangers in the CVCS. The EPRI survey indicatedoperational practices could be optimized.

The EPRI survey indicated that 78% of the domestic plants were using gel-type cationresin, with no clear benefit in using macroreticular-type resins, and that 75% of thedomestic PWRs indicated that low chloride (<0.1%) anion resin was being used in themixed beds. Resin particle size or other attributes such as crosslinkage were notreported to have a significant effect on performance. However, the traditional practiceof using a 1:1 equivalent cation-to-anion mix in the mixed beds is not always beingmaintained by domestic or German PWRs. Customized blends to address particular


D-4

purification requirements were reported, with various resin ratios being useddepending on particular needs.

Three practices are used to convert the hydrogen-form cation to the Li-7 form:

• The resin is purchased in the Li-7 form.

• The resin is lithiated in situ by Li7OH additions and flushed to waste.

• The spare bed is valved in as needed to maintain reactor coolant lithium levelswithin the control band (delithiation), and this bed is used during the subsequentfuel cycle (i.e., lithium management is practiced).

The EPRI survey indicated that the above practices could be modified to optimize theresin usage and potentially minimize lithium additions as follows, depending on plantdesign:

• Plants with deborating beds could partially fill one vessel with anion resin and theremaining vessels charged fully with anion resin for deboration at the end of thefuel cycle. The partially filled vessel can be filled with cation resin in the hydrogenform for complete lithium removal in the hot boration step (acid-reducingchemistry) for crud removal.

• Plants without deborating beds can use one vessel in the Li7-borate form and thesecond bed in the hydrogen-borate form, with a resin mix higher than the normal1:1 cation-to-anion mix. This bed is valved in as needed for lithium control. At theend of the fuel cycle, the second bed is used for complete lithium removal, and sinceexcess cation capacity in the hydrogen form exists, this bed is also used forcobalt/nickel removal. The original lithiated bed is returned to service andmonitored for performance. Since this bed was not used during crud dissolutionsteps, it will have a relatively low nickel inventory which can be subsequentlyreleased and activated to Co-58. If this practice is conducted in plants with separatecation beds, the cation beds potentially can be used for several fuel cycles.

• Near the end of the fuel cycle a mixed bed in the hydrogen-hydroxide form islithiated but not borated. This bed is now used for boron reductions as needed.After deboration this bed is placed in service full time and the original mixed bed isremoved from service for the power coastdown. Full boration of the new bedoccurs upon boration for shutdown, at which time the bed is removed from serviceand replaced with the original bed plus the cation bed. This train is now used forlithium removal and crud/nickel removal. Upon startup, the bed installed duringthe coastdown is placed in service as the normal purification bed. A new cation bedis valved in as needed for lithium control.


D-5

Plants responding to the EPRI survey reported few problems with resin performance:

• Lithium control required frequent bed alignments for lithium removal.

• Sulfate spikes occurred during shutdown with and without hydrogen peroxideadditions. The presence of leachable species containing sulfur from a naked cationbed in the hydrogen form was a potential concern.

• Chloride and fluoride spikes were reported.

• One plant reported high H-3 levels attributed to 96.6% Li-7 in a purchased lithiatedbed.

• One plant reported low iodine DF during hydrogen peroxide addition.

The main conclusion from the EPRI survey is that CVCS resin can be optimized byusing different cation-to-anion resin mixes in the mixed beds, layering resin beds tooptimize performance and implementing various lithium management practices tooptimize bed life. A potential area for improvement reported was the use of reverseosmosis for lithium and boron recovery and specialized resins to target specificradionuclides.

A recent EPRI sponsored study indicated that the large costs associated with resindisposal are an important factor in resin/ion exchange practice optimization, and canpromote use of segregated beds versus mixed beds (2).

4.0 References

1. J. C. Bates and M. E. Pick, "CVCS Resin Management Strategies Procedures; Optionsfor the Sizewell B Design," Water Chemistry of Nuclear Reactor Systems 7. BNES,Vol. 1, 81, 1996.

2. Analysis of Advanced Liquid Waste Minimization Techniques at a PWR. ElectricPower Research Institute, Palo Alto, CA. April 1998. TR-109444.

E-1

E OXYGEN AND HYDROGEN BEHAVIOR IN PWR

PRIMARY CIRCUITS

1.0 Summary

Oxygen and hydrogen behavior in a PWR primary circuit are linked by the radiolysisreactions occurring in the core to the extent that they cannot normally co-exist in thecoolant, other than upstream of the core. In normal operation and during shutdownthe important aspects are – (1) what is the minimum hydrogen level to suppressradiolysis in the water phase in the core, (2) will sub-cooled nucleate boiling on fuelassemblies in the core deplete the hydrogen in the water phase below this minimumhydrogen concentration and (3) what is the effect of oxygen ingress into the RCS.

The concentrations of oxygen, hydrogen, hydrogen peroxide and other species in aPWR primary circuit are controlled by approximately 50 radiolytic and thermalreactions, all of which occur simultaneously as the coolant circulates around the circuit.The rate constants of these reactions are known over the full operating temperaturerange of the primary circuit and the behavior can be modeled with reasonableconfidence. This appendix describes the overall behavior and is based on Refs. 1 and 2,which model oxygen/hydrogen concentrations in a typical Westinghouse 4-loop PWR

(Subsequent to preparation of this appendix and the two reports on which it is based (1,2), EPRI has become aware of recent work that indicates that some factors notaddressed in this appendix, e.g., enhanced radiolysis due to the effects of impurities,may need to be considered (12). It is suggested that (12) be considered in anyevaluations of the routine use of hydrogen levels below those specified in Table 3-3.)

2.0 Radiation Chemistry

Radiolysis decomposes water into the relatively stable molecular species H2 and H2O2

and the radicals e-aq, H⋅ and OH⋅, with the yields being dependent on the linear energy

transfer (LET) of the radiation. Of the three types of radiation present in an operatingPWR, γ-rays are a low LET radiation giving relatively high radical yields and lowmolecular product yields, fast neutrons are an intermediate LET radiation with lower

Oxygen and Hydrogen Behavior in PWR Primary Circuits

E-2

radical yields and higher molecular yields and the 10B(n,α)7Li reaction products givehigh LET radiation producing low radical yields and high molecular yields. Themolecular species and radicals formed then interact with each other to form O2 and O2

-

in reactions such as,

OH⋅ + H2O2 → HO2⋅ + H2OHO2⋅ + HO2⋅ → O2 + H2O2

and e-(aq) + O2 → O2

-

Hydrogen is added to prevent oxidizing conditions from existing in the primary circuit.When oxygen is present in sufficient quantities, hydrogen causes a chain reaction tooccur, which promotes H2O2 (and O2) decomposition via the following reactions,

OH⋅ + H2 → H⋅ + H2OH⋅ + H2O2 → OH⋅ + H2O

and e-(aq) + H2O2 → OH- + OH⋅

for which terminating reactions are,

OH⋅ + H2O2 → HO2⋅ + H2Oand O2 + H⋅ →HO2⋅

With excess H2 present, the net result is that only small steady-state concentrations ofO2, O2

-, HO2⋅ and H2O2 exist, with the concentrations being controlled by the rates ofproduction and destruction for each species. In the presence of excess O2 the destructionreactions are suppressed and radiolytic decomposition to O2, O2

-, HO2⋅ and H2O2 isfavored, resulting in much higher steady state concentrations and the generation ofoxidizing conditions.

At near ambient temperatures the radical and molecular yields for the different types ofradiation are such that the minimum hydrogen concentration that would be required tosuppress radiolysis in a PWR during power operation is 10 to 15 cc (STP) kg-1, a valuethat has been in use since the 1960s. However, the minimum hydrogen concentration tosuppress radiolysis is very temperature dependent and reduces as the operatingtemperature increases. This is mainly due to the temperature dependence of thereaction,

OH⋅ + H2 → H⋅ + H2O

which at high temperature increases the production rate of H⋅. This in turn reacts toform e-

(aq), which reacts with H2O2 reducing its steady state concentration as thetemperature increases and stops radical-radical reactions, which would tend to lead tonet radiolytic decomposition of water to hydrogen peroxide and oxygen.


E-3

3.0 Minimum Hydrogen Concentrations at Full Power

The results of the latest modeling studies are shown in Figs. E-1 and E-2. For bothstart-of-cycle (1800 ppm boron) and end-of-cycle (0 ppm boron) conditions, thebehavior around the primary circuit is similar in all cases, e.g., Fig. E-1.

For all conditions examined O2, O2-, HO2⋅ and H2O2 concentrations reach steady-state

values in the first few centimeters (<10 cm) of the core. Out-of-core the radical andmolecular concentrations then decrease rapidly by 1 to 2 orders of magnitude. Forexample O2 and H2O2 react with H⋅, OH⋅ and e-

(aq), whilst O2- and HO2⋅ react with other

radicals (or HO2⋅ with itself) to form neutral species.

For both start-of-cycle (1800 ppm boron) and end-of-cycle (0 ppm boron) conditions at300°C (572°F), the calculations predict that similar minimum hydrogen concentrationsare required to suppress radiolysis at all boron concentrations, see Figs. E-2 and E-3,with the main difference being that the steady-state O2 and H2O2 concentrations at end-of-cycle will be between two and ten times lower. Below the minimum hydrogenconcentration radiolytic decomposition gives high in-core steady-state H2O2 and O2

concentrations, which only reduce marginally out-of-core. Adding only a small amount

1.00E-17

1.00E-16

1.00E-15

1.00E-14

1.00E-13

1.00E-12

1.00E-11

1.00E-10

1.00E-09

1.00E-08

1.00E-07

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1.00E+00

0 20 40 60 80 100 120

Distance/m

Conc

entra

tions

/ppm

Exits core

Start ofSteamGenerator

Start of Cold leg

O2

H2O2

Enters core Exits core

CVCS enters main circuit

Start ofCrossoverleg

Figure E-1Concentrations of O2 and H2O2 Around the Main Reactor Loop During Full PowerOperation. 25 cc/kg H2, 1800 ppm Boron. Two Transits Around the Main Circuit areShown (taken as 56 m long).


E-4

1.00E-11

1.00E-10

1.00E-09

1.00E-08

1.00E-07

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1.00E+00

1.00E+01

1.00E+02

1.00E+03

0.1 1 10 100H2 (cc/kg)

Con

cent

ratio

n/pp

m

O2 In-core

O2 Downstream ofcore

H2O2 Downstream ofcore

H2O2 In-core

Figure E-2Steady-state O2 and H2O2 Concentrations at 1800 ppm Boron, Corresponding tothe Start of an 18 Month Fuel Cycle (pH(300 C)6.9)

1.00E-12

1.00E-11

1.00E-10

1.00E-09

1.00E-08

1.00E-07

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1.00E+00

1.00E+01

1.00E+02

0.1 1 10 100

H2 (cc/kg)

Con

cent

ratio

n/pp

m

O2 In-core

O2 Downstream ofcore

H2O2 Downstream ofcore

H2O2 In-core

Figure E-3Steady-state O2 and H2O2 Concentrations at 0 ppm Boron, Corresponding to theEnd of an 18 Month Fuel Cycle (pH(300 C)7.4)


E-5

of H2 suppresses radiolytic decomposition and this is predicted to be essentiallycomplete when the concentration reaches 0.5 cc (STP) kg-1 H2. Increasing the hydrogenconcentration further has only a small effect on the H2O2 level, but oxygen continues tofall. Although not shown, the concentrations of O2

- and the HO2⋅ radical are predicted tobe small at all H2 concentrations (although O2

- is higher than the O2 concentration) andwould not be expected to have a significant impact on crud behavior in-core. In-corethe OH⋅ radical, which is a powerful oxidant in its own right, can have significantconcentrations, particularly at intermediate hydrogen concentrations.

Whilst the lower steady state concentrations of O2 and H2O2 at end-of-cycle areexpected, the similarity in the minimum hydrogen concentrations to suppressradiolysis at start and end-of-cycle has not previously been predicted. Burns (Ref. 3),however, reported that the minimum hydrogen concentration is proportional to theratio of H2 to OH⋅ production. At 300°C this ratio is 0.5 (γ + n) for end-of-cycleconditions (0 ppm boron) and 0.77 (γ + n + α) for start-of-cycle conditions (1800 ppmboron), suggesting that any boric acid effect will be small. The results confirm that theminimum hydrogen concentrations differ by a factor of <2.

These very low thresholds are consistent with the experimental results obtained byboth McCracken (~0.3 cc (STP) kg-1, Ref. 4) and Pastina (<1 cc (STP) kg-1, Refs. 5 and 6).Christensen (Refs. 7 and 8) has also carried out modeling studies, the latest of whichalso predicts that radiolytic decomposition only occurs at <0.3 cc (STP) kg-1. The resultsare also consistent with studies carried out at power using the Belleville 1300 MWePWR, where some radiolytic decomposition was inferred when the coolant hydrogenlevels were reduced to between 3.6 and 5 cc (STP) kg-1 (Ref. 9) and in the Bruce CANDUreactor where the inferred minimum deuterium concentration to suppress radiolysiswas 1.3 cc (STP) kg-1 D2 (Ref. 10).

4.0 The Effect of Voidage on Minimum Hydrogen Levels

In Revision 3 of the Guidelines, it was suggested that operating in the upper half of thedissolved hydrogen band might help to avoid heavy fuel crudding and hence, AxialOffset Anomalies. More recently it has been suggested that when sub-cooled nucleateboiling occurs on the highest rated assemblies, the H2 concentration will be reducedlocally to such an extent that O2 and H2O2 may be formed. In-core this could reducecorrosion product solubilities, leading to enhanced crud deposition as observed underboiling conditions. Downstream of the core, if any O2 and H2O2 survive to reach theSGs, they might also increase the corrosion release rate of Ni from Inconel 600/690 SGtubing and this could explain the relatively high Ni/Fe ratios in crud on boiling fuelassemblies.

Assuming that equilibrium conditions are maintained under sub-cooled nucleate boilingconditions, then from the appropriate mass balance equations in the presence of a steam


E-6

bubble it is easy to show (Ref. 1) that the concentration of hydrogen in the liquid in theabsence of voidage C0 and the concentration in the presence of the void, Cl, are relatedby:

F) - (1PF + 1

F) - (1

F + 1

= CC

V

l

g

0

l ρ

ρ

where F is the volume void fraction, gρ is the gas density, lρ the liquid density and PV isthe volume partition coefficient (PV = Cl/Cg).

To produce an estimate of the effect of sub-cooled nucleate boiling it is necessary toassume a voidage level for the highest rated fuel assembly. Using a value of 10%, i.e.,F = 0.1, which is higher than the maximum in most cores, gives,

Cl /C0 = 0.795

indicating that even at this high voidage level 80% of the dissolved hydrogen remainsin the water phase. Thus, even if the dissolved hydrogen level was at the bottom of therecommended range, i.e., 25 cc (STP) kg-1, and if the hydrogen has time to partition toequilibrium in the bubbles, the mean concentration of hydrogen in the water at thepoint of maximum void fraction would only decrease to ~20 cc (STP) kg-1. This is stillsufficient to suppress radiolysis and is greatly in excess of the minimum hydrogenconcentration.

An alternative treatment by Burrill (Ref. 10) is based on the assumption that the growthof a steam bubble reduces the hydrogen concentration locally in the water layersurrounding the bubble. Excluding diffusion in the water phase, this treatment predictsthat the local concentration will be reduced to a value close to the minimum hydrogenconcentration, suggesting that some radiolysis could occur even when the bulkconcentration is in the normal operating range. However, in a more detailed treatmentby Zmitko (Ref. 11) for VVER reactors, which included diffusion from the liquid phase,depletion of the water phase around the steam bubble was less complete (about 70%),which would still maintain sufficient hydrogen in the water phase to suppressradiolysis.

5.0 Oxygen Ingress

The addition of reactor make-up water, concentrated boric acid and/or blended boricacid via the CVCS charging line adds oxygen to the RCS. Oxygen addition can occurunder a variety of circumstances and raises the questions of what parts of the primary


E-7

circuit and CVCS pipework are affected by oxygen ingress and what are theconsequences of the addition. Some examples are listed below,

1. Oxygen ingress during load following. This is known to occur during loadfollowing, which initially requires adding boric acid to reduce power load, followedby dilution with reactor make-up water to return to 100% power. Oxygen addedwith the reactor make-up water has been suggested to be the cause of azimuthaloffset anomalies and an increase in 51Cr inventories in some load following EDF1300 MWe PWRs (13).

2. Oxygen ingress at power towards end-of-cycle. For PWRs without a BTRS,increasingly larger volumes of reactor make-up water are added at end-of-cycle todilute the boron concentration, carrying with it the risk of significant oxygen ingressinto the RCS.

3. Oxygen ingress at power. A number of stations have fully oxygenated reactormake-up water tanks. In VVERs a large fraction of the make-up enters the primarycircuit via the main coolant pump seals. In these stations, which only have stainlesssteel in the RCS, it has been reported that the oxides adjacent to the RCPs areprimarily hematite, rather than the normal spinels grown under reducingconditions.

4. Oxygen ingress during boration under hot shutdown conditions. Under theseconditions part or all of the CVCS charging and seal injection flows can beoxygenated, with the CVCS oxygen concentration ranging from a fraction of a ppmup to fully aerated.

5. Oxygen ingress on aligning the RHRS trains at <350°F (<177°C). Depending on thestate of the RHRS trains, the coolant in the RHRS trains could have an oxygenconcentration ranging from zero to fully aerated. Before use the RHRS trains mustfirst be warmed through to avoid thermal shock. For this, a low flow from the RCShot leg is passed through the RHRS trains, normally routed into the CVCS via theLP letdown connection and all oxygenated water will probably be purged to theCVCS before the RHRS trains are aligned for cooling duty. The CVCS charging flowmay contain oxygen ranging in concentration from a fraction of a ppm to severalppm. Some stations add hydrazine at this stage, either to the RHRS or the RCS tomitigate against oxygen addition, whilst others route the warm-through RHRS flowto the boron recycle system and replace it with blended boric acid having a lowerdissolved oxygen content.

6. Oxygen ingress on adding blended boric acid during cooldown to accommodatechanges in density and on collapsing the pressuriser bubble. CVCS charging flowconditions will be similar to (4) and (5) above.


E-8

7. Oxygen ingress into the pressuriser during shutdown. In principle, boric acid andblended boric acid can be added to the RCS via the auxiliary spray line and localoxygenated conditions can be formed in the pressuriser.

8. Oxygen removal during start-up. Normally the RCS is fully oxygenated beforestart-up and the O2 concentration is reduced by adding hydrazine, via the CVCS,before the temperature can be raised above 250°F (120°C, <100 ppb oxygen). If thesprays are not fully opened to maximize the flow rate through the pressuriser, thelatter could remain static and oxygenated up to higher temperatures than normallypermitted. Some PWRs avoid this situation by adding hydrazine to the pressuriserand reducing its oxygen concentration before pressuriser heat-up commences.Commonwealth Edison PWRs add hydrogen, rather than hydrazine, to removeoxygen during start-up.

The effects of oxygen ingress can be understood by looking at the processes that controlthe variation in the oxygen concentration around the RCS and CVCS at particular timesduring power operation/shutdown/start-up. The main processes affecting theconcentration are the radiolytic reactions occurring in the core, reactions with surfaceoxides and thermal and radiolytic reactions with additives such as hydrazine.Although the oxygen concentration in the CVCS charging line can be fully aerated, itwill always be very greatly diluted when it is added to the RCS because of the veryhigh flow rates in the primary circuit. The actual concentration in the RCS will dependon the CVCS and RCS flow rates, but typically a concentration of 6 ppm dissolved O2 inthe CVCS charging line will be reduced to 4 – 10 ppb in the RCS cold leg immediatelydownstream of the CVCS return flow. In all instances good mixing should occur, due tothe very turbulent conditions existing in the RCS.

The behavior when aerated water is added from the CVCS at full power is shown inFig. E-4. Within the core the oxygen reacts radiolytically very rapidly with the excesshydrogen in the primary coolant and essentially all reacts in the first few centimeters(<10 cm) of the core. Although H2O2 is produced, and is the dominant species present,its concentration remains low. Beyond the core the concentrations reduce rapidly by 1to 2 orders of magnitude. Whilst O2

- and HO2⋅ are also produced in-core theirconcentrations are low and they will rapidly disappear from the primary coolantbeyond the core. Thus, effectively, the concentrations of all oxidants become negligibledownstream of the core.

Comparison of these results with those obtained when there is no oxygen ingress showsthat the steady-state in-core and downstream out-of-core concentrations of all speciesare essentially identical. Hence, any effect due to oxygen ingress must be restricted tothe section of the single cold leg downstream of the CVCS charging line return andmainly to that part of the RPV downcomer and lower plenum that this cold legsupplies.


E-9

During a shutdown the behavior is generally similar to that at full power and H2O2

remains the dominant oxidizing species present, although the steady state levels of bothO2 and H2O2 alter as the temperature and γ-dose rate decreases. Under all conditions,increasing the H2 level suppresses the steady-state O2 concentration, but not the H2O2

concentration, which remains essentially constant once a small H2 excess is present.Oxygen ingress also has a similar effect, with the steady-state in-core and downstreamout-of-core concentrations being essentially identical with or without oxygen ingress,but with the radiolytic equilibrium occurring progressively further up the core as theshutdown γ-radiation field decays. As in the full power case, oxygen ingress can onlyaffect the section of the cold leg downstream of the CVCS charging line return and thepart of the RPV that this cold leg supplies.

1.00E-17

1.00E-16

1.00E-15

1.00E-14

1.00E-13

1.00E-12

1.00E-11

1.00E-10

1.00E-09

1.00E-08

1.00E-07

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1.00E+00

0 20 40 60 80 100 120Distance/m

Conc

entra

tions

/ppm

CVCS Enters main loop Enters reactor core

Exits core

Exits core

Start ofSteamGenerator

Start ofCrossoverleg

Start of Cold leg

O2

H2O2

Exits core

Figure E-4Concentrations of O2 and H2O2 Around the Main Reactor Loop. 25cc/kg H2, 1800ppm B, Full Power Operation, Oxygen Ingress from the CVCS. Two TransitsAround the Main Circuit are Shown (taken as 56 m long).

During hot shutdown, at up to eight hours after shutdown, the calculations predict thatO2 entering the core will react very rapidly with excess H2 in the coolant and will beremoved in the first ten centimeters of the core. Downstream oxygen levels always fallby about two orders of magnitude, but after eight hours H2O2 only falls by a factor ofabout 2. The latter is due mainly to the reduced OH⋅ concentration because of the muchlower γ-dose rate. The concentrations then remain low until oxygen is again added inthe cold leg. Comparing the results with those obtained in the absence of oxygeningress, shows that, with excess H2 present, the steady-state concentrations of all species


E-10

are essentially identical. Hence, the only part of the circuit affected by the addition ofaerated water is the cold leg downstream of the CVCS charging line.

For RHRS alignment and hydrogen peroxide addition (assumed to occur after 24 and48 hours), modeling shows that the oxygen still reacts very rapidly with excess H2 inthe primary coolant in the lower part of the core, Fig. E-5. For a hydrogen concentrationof 5 cc (STP) kg-1 the H2O2 concentration increases to about 1 ppb on RHRS alignmentand to 12 ppb immediately before H2O2 addition, whereas O2 remains very low.Furthermore, whilst the O2 concentration still decreases by about one order ofmagnitude downstream of the core at both 350°F and 120°F, the H2O2 concentrationbecomes increasingly less affected and by 120°F, immediately before H2O2 addition,hardly changes downstream of the core. The consequence will be that hydrogenperoxide will start to penetrate around the remainder of the primary circuit as the coreγ-dose rate and temperature decreases, although at the higher temperatures the effectwill be smaller and would be retarded by loss of H2O2 by heterogeneous reaction withthe oxide covered surfaces of the primary circuit.

1 .00E -10

1 .00E -09

1 .00E -08

1 .00E -07

1 .00E -06

1 .00E -05

1 .00E -04

1 .00E -03

1 .00E -02

1 .00E -01

1 .00E + 00

0 0 .5 1 1 .5 2 2 .5 3 3 .5 4D ista n ce /m

Con

cent

ratio

n/pp

m 10x s to ic hiom etry H210x s to ic hiom etry H25 cc /kg H25 cc /kg H2

O 2

H 2O 2

Figure E-5Variation of O2 and H2O2 Concentrations Through the Core for Different HydrogenLevels, O2 Ingress, 1 Day After Shut down, 177 C.

To avoid the effects of oxygen ingress some stations add hydrazine, either to the RHRStrains or directly to the RCS. Hydrazine reacts with oxygen in aqueous solution bythermal and radiolytic reactions and by surface heterogeneous reactions. Both the


E-11

thermal and radiolytic reactions are known to proceed via the neutral species N2H4 andN2H3⋅, rather than the protonated species N2H5

+ and N2H4⋅+ (pKa 8.1 and 7.2 at 25°C,

respectively). In the primary circuit the radiolytic reactions are dominant and occurvery rapidly in the core. Compared with the radiolytic reactions, the thermal andheterogeneous reactions are relatively slow, are not well characterized and arecomplicated by effects of metal ions and surfaces, both of which catalyze the reactions(Ref. 1). Because of the relative reaction rates it is clear that if hydrazine is added to theRCS directly, even at 350°F the thermal reaction rate is not rapid enough to removeoxygen in the short time interval (approximately 1 sec) between the oxygen beingadded to one of the cold legs via the CVCS charging line and the coolant reaching thecore. As a result, the primary oxygen removal mechanism must be the radiolyticreaction in the core with dissolved hydrogen in the primary coolant and hydrazineaddition to the RCS can be of no significant benefit. Although hydrazine cannotdeoxygenate the RCS, it will be capable of removing oxygen from RHRS trainsthemselves. At about 85ºF the half lives for oxygen removal are expected to be long(particularly recognizing that the rates will be slower under acid conditions) and fulldeoxygenation may take several months.

6.0 Start-up Deoxygenation

At start-up deoxygenation is normally achieved by the addition of hydrazine to theRCS. As described above, the thermal reaction rates are relatively slow, but increasewith temperature, whilst the radiolytic rate will be dependent on the core γ-radiationdose rate. At start-up the latter will be dependent on how many new fuel assembliesare loaded and the time since the reactor shut down. Oxygen can also be removed byreaction with oxides grown under reducing conditions, since such oxides are known tohave significant oxygen capacities, but these are relatively unimportant at start-uptemperatures.

As outlined above, hydrazine reacts with oxygen by both thermal and radiolyticreactions, with the thermal reactions including both homogeneous solution andheterogeneous surface reactions and with some reactions being catalyzed by metal ions,particularly copper. With the exception of the pressurizer, under all conditions theradiolytic reactions will be dominant. However, from Ref. 1, the half-life of the thermalhomogeneous reaction decreases rapidly between 120°F and 350°F and, if practicable, itclearly is always an advantage to deoxygenate the RCS at higher temperatures. It mayalso be noted that reaction rates will be less under the acidic conditions existing in theRCS, but that this may be counteracted by surface and metal ion catalyzed reactions.

The radiolytic reaction of hydrazine and oxygen in the core is also complex andinvolves a chain reaction at pH >7, but at lower pH hydrazine is decomposed with ayield of ~2 molecules/100 eV at room temperature through a set of reactions, which donot include a chain reaction, initiated by reaction of OH⋅ with N2H4,


E-12

N2H4 + OH⋅ → N2H3⋅ + H2O

followed by,

N2H3⋅ + O2 → N2H2 + HO2⋅

which is analogous to the case with H2. The subsequent reactions are unclear, but thereis significant decomposition of N2H4 and H2O2 could be one of the products.

Deoxygenation can also be carried out by adding hydrogen to the RCS via the VCT. Inthis case hydrogen will not react thermally with oxygen, but it will remove oxygenefficiently by radiolytic reactions in the core. Calculations show that after refueling,even several months after shutdown, a sufficient γ-dose rate will exist in the core todrive the reaction, provided the γ-dose rate remains above about 1 MRad h-1.Deoxygenation is relatively insensitive to RHRS and pressuriser flow rates, providedthere is sufficient flow to give an adequate turnover of their contents. However, it issensitive to the rate of H2 supply into the RCS, which in turn is dependent on the CVCSflow rate, VCT purge flow rate and water level in the VCT (which should be high tominimize the gas space volume). Burping the VCT is very efficient at changing the VCTgas space volume from N2 containing some O2 to pure H2, which in turn has a markedeffect on reducing the overall deoxygenation time.

Compared with thermal and radiolytic reactions of oxygen with hydrazine, radiolyticdeoxygenation using hydrogen is a more efficient way of removing oxygen from theprimary coolant. Similar radiolytic reactions occur with N2H4, but they are somewhatmore complex and the process is almost certainly less efficient and in acid solution atstart-up the reaction with N2H4 is unlikely to be as rapid as with hydrogen, because theradical N2H4⋅

+ is thought not to react with oxygen. Potentially, hydrogen has thedrawback that it does not react thermally with oxygen and that explosive mixturescould be formed in the VCT. However, in practice, there should always be a sufficientγ-dose rate to drive the radiolytic reactions, whilst calculations show that, for a VCTpurged with hydrogen, the oxygen concentrations in the VCT are always too low toform explosive mixture.

7.0 Spent Fuel Pool

In addition to radiolysis determining the behavior of oxygen and hydrogen species inthe primary circuit, the residual γ-dose rate of the fuel assemblies will control thespeciation in the spent fuel pool. In the spent fuel pool the water will always besaturated with respect to oxygen, but hydrogen peroxide will also form and will buildup to significant levels. Although few measurements have been reported, theconcentrations in Spanish PWRs and at Sizewell B are normally about 4 ppm H2O2 andthis is probably typical for all PWRs.


E-13

8.0 Conclusions

Overall, the general conclusions of the experimental and modeling studies are,

1. There is clear agreement from experimental and modeling data that the dissolvedhydrogen concentration required to suppress radiolysis decreases as thetemperature increases. At PWR operating temperatures it is much lower than thecurrently accepted value of 10 to 15 cc (STP) kg-1 and is in the range <1 to 5 cc (STP)kg-1 hydrogen. Reactor data suggest that the minimum hydrogen level is ~5 cc (STP)kg-1 H2 during power operation, whilst modeling indicates that the minimumhydrogen concentration is about 0.5 cc (STP) kg-1 at PWR operating temperatures.The presence of boric acid has little effect on this value, although it does increase thesteady-state concentrations of O2 and H2O2.

2. Assuming that equilibrium is achieved, sub-cooled nucleate boiling will have only arelatively small effect on the hydrogen concentrations in the water phase in thehighest rated fuel assemblies and should not reduce concentrations to the pointwhere radiolysis will occur to produce increased concentrations of hydrogenperoxide and oxygen.

3. Under all conditions the dominant oxidizing species formed in-core is hydrogenperoxide, which can be present at levels several orders of magnitude higher thanoxygen. At shut down the in-core steady-state hydrogen peroxide concentrationsincrease significantly as the core γ-dose rates and temperature reduce, whilst thereduction in concentration downstream of the core also becomes less. The increasein hydrogen peroxide concentrations is caused by the reduction in primary circuittemperature and will occur both with or without oxygen ingress into the primarycircuit.

4. In the primary circuit, any effects upstream of the core from oxygen ingress will belimited to the cold leg of the loop receiving the CVCS charging line flow and,mainly, to the quadrant of the RPV and core that this loop feeds. At power or hotshutdown, oxidant concentrations downstream of the core will be effectively zeroand any effect on the circuit materials will be negligible. At lower temperatureshydrogen peroxide will exist downstream of the core, although its appearance willbe retarded by decomposition reactions on the oxide covered surfaces of theprimary circuit.

5. The use of hydrazine to counteract the effects of any oxygen present in the RHRStrains will only be effective if it is added to the RHRS trains, with sufficient timeallowed for it to slowly deoxygenate the system before a shutdown. At start-up,deoxygenation of the primary coolant with hydrazine proceeds via both thermaland radiolytic reactions. Deoxygenation can also be achieved radiolytically usinghydrogen. Hydrogen is expected to react more efficiently than hydrazine in an acid


E-14

medium.(Subsequent to preparation of this appendix and the two reports on which it isbased (1, 2), EPRI has become aware of recent work that indicates that some factorsnot addressed in this appendix, e.g., enhanced radiolysis due to the effects ofimpurities, may need to be considered (12). It is suggested that (12) be consideredin any evaluations of the routine use of hydrogen levels below those specified inTable 3-3.)

9.0 References

1. Oxygen and Hydrogen Behavior in PWR Primary Circuits. Electric Power ResearchInstitute, Palo Alto, CA. _______, 1999. TR-XXXXX (in course of publication).

2. K Garbett, J Henshaw, M V Polley and H E Sims, “Oxygen and Hydrogen Behaviorin PWR Primary Circuits”, 1998 JAIF Int. Conf. on Water Chemistry in NuclearPower Plants, Kashiwazaki City, Japan, October, 902, 1998.

3. W G Burns, Suppression by Dissolved Hydrogen of the Radiolysis of Water byMixed Radiation Fields, UKAEA Report, AERE-M 2702, 1975.

4. D R McCracken, Coolant Radiolysis and Boiling in Water-Cooled Reactors, EPRIReport TR-100789, 1992.

5. B Pastina, Etude sur la Radiolyse de l’Eau en Relation avec le Circuit Primaire deRefroidissement des Réacteurs Nucléaires à Eau Sous Pression, PhD ThesisUniversity of Paris, XI Orsay, 1997.

6. B Pastina, J Isabey and B Hickel, “Water Radiolysis: the Influence of Some RelevantParameters in PWR Nuclear Reactors”, Water Chemistry of Nuclear ReactorSystems 7, BNES, London, Vol 1, 153, 1996.

7. H Christensen, “Remodelling of the Oxidant Species during Radiolysis of HighTemperature Water in a Pressurised Water Reactor”, Nuclear Technology, 373, 109,1994.

8. K Hisamune, M Sekiguchi, H Takiguchi and H Christensen, “New Aspect of DHControl in PWR Primary Water Chemistry”, 1998 JAIF Int. Conf. on WaterChemistry in Nuclear Power Plants, Kashiwazaki City, Japan, October, 595, 1998.

9. C Brun, A Long, P Saurin, M C Thiry and N Lacoudre, “Radiolysis Studies atBelleville PWR - Water Chemistry with Low Hydrogen Concentration, Chemistry inWater Reactors: Operating Experience and New Developments”, European NuclearSociety Vol 1, 73 to 79, 1994.


E-15

10. K Burrill, “Some Aspects of Water Chemistry in the CANDU Primary WaterCoolant Circuit”, 1998 JAIF Int. Conf. on Water Chemistry in Nuclear Power Plants,Kashiwazaki City, Japan, October, 426, 1998.

11. M Zmitko, V Linek and T Moucha, “Subcooled Boiling Effect on Dissolved GasesBehavior, Water Chemistry and Corrosion Control of Cladding and Primary CircuitComponents”, IAEA Technical Committee Meeting, Hluboká n. Vltavou, CzechRepublic, September, 1998.

12. D. R. McCracken, et al., Aspects of the Physics and Chemistry of Water Radiolysisby Fast Neutrons and Fast Electrons in Nuclear Reactors, Report AECL-11895,AECL, Mississauga, Ontario, September 1998.

13. R. Roofthooft, et al., Primary Circuit Chemistry of Western PWR and VVER PowerPlants, Unipede report 02004Ren9653, p. 69, Unipede, Paris, 1996.

F-1

F CONSIDERATIONS FOR REFERENCING HIGH

TEMPERATURE PH CALCULATIONS TO 300°C

1.0 Introduction

The current practice in the U.S. relative to control of RCS pHT is to base the calculationon the average core temperature (Tave). In the European community, calculations arebased on a constant temperature of 300°C regardless of plant operating temperatures.For plants with Tave>300°C, less lithium is required to achieve a given pHT if thecalculation is based on Tave rather than 300°C. For plants with Tave<300°C, the reverse istrue.

In the past several revisions of these Guidelines, the average core temperature wasselected as a basis of the pHT calculation to reflect plant design characteristics. Inparticular, the original goal was to set a pHTave such that the solubility of Fe3O4 had aminimum near the center of the core. The reason for setting it at the core midpoint wasthat deposition generally predominated in the top half of the core. Although the ironsolubility from magnetite was a minimum at pH300°C = 6.9, selection of the average coretemperature was based on the observation that the shape of the solubility curveremained reasonably similar in the core region although the absolute solubilityincreased or decreased as T varied if control was based on pHTave. The same can be saidfor variations in iron solubility from nickel ferrite across various cores when a pHTave

calculational basis is employed.

Calculation of pH based on a fixed reference temperature, like 300°C, is founded on theobservation that the temperature dependence of pH is principally controlled by thestrong variation of the dissociation constant of water, Kw, with temperature. The pOHacross a typical core, in fact, changes little from inlet to outlet. Fixing the temperatureat which the pH is calculated has the equivalent effect as using pOH for lithium-boroncontrol in that it eliminates lithium addition and removal operations which otherwisewould be demanded solely by the sensitivity of Kw to temperature. Using a referencetemperature also can facilitate comparison of RCS chemistry environments acrossdifferent cycles and at different plants. The choice of reference temperature is

Considerations for Referencing High Temperature PH Calculations to 300 C

F-2

arbitrary, with the argument in favor of 300°C usually deriving from similarobservations about corrosion product solubility as noted in the Tave case.

Since the temperature on which the pHT calculation is based impacts on the requiredlithium concentration and thus on corrosion rates, on corrosion product transport andon activity transport, past practices and solution chemistry changes associated withselecting 300°C versus Tave for the pHT calculation are summarized below. Thisdiscussion is particularly relevant at this writing as more plants are running highertemperature cores over extended cycles. The difference in early cycle lithiumcalculated using 300°C versus Tave for increasing Tave can become significant in regard tocorrosion product release to the coolant and crud deposition on the core.

2.0 Background

Coordinated chemistry control was originally based on early laboratory studies (1) ofmagnetite solubility using molal base concentrations at 285°C (Tc) and also usingdifferent lithium-to-boron ratios as a function of temperature (2). This work wasexpanded to include the minimum lithium for a zero temperature coefficient ofsolubility of nickel ferrite as Ni0.5Fe2.5O4 at 293°C (a more representative Tc) as a functionof boron concentration. The lithium at the beginning of a conventional fuel cycle (1200ppm boron) to provide the zero temperature coefficient of solubility was approximately7.5 ppm, which was well above the maximum allowed concentration based on fuelcorrosion considerations (2.2 ppm lithium has historically been accepted as a safe limitfor non-boiling cores). The room temperature pH of the solutions in these early studieswas occasionally reported, but pH at temperature was not commonly used.

The first EPRI PWR Primary Water Chemistry Guidelines issued in 1986 (3) recommendedcoordinating lithium with boron (based on magnetite solubility at 285°C to provide aminimum pH of 6.8), but permitted operation with "elevated lithium" during the cycle."Elevated lithium" at this time was operation in a range between 2.2 ppm and the lowerboundary for coordinated lithium, with lithium limits reduced between approximately225 ppm boron and the end of the cycle. The Guidelines indicated that the lithiumreduction was based on maintaining a high temperature pH of 7.5, which waspredicted by the CRUDSIM model to achieve minimum deposition of nickel ferrite onthe core. The CRUDSIM model employs iron solubility data to predict core-wide crudbuild-up for specified nickel stoichiometry in the ferrite deposits.

PWR Primary Water Chemistry Guidelines: Revision 1 issued in 1988 (4) indicated thatoperation at constant pHT of 7.4 or above during the fuel cycle would minimize thebuildup of activity on out-of-core surfaces. The Guidelines revision also indicated thatCRUDSIM predicted that maintaining a constant 2.2 ppm lithium during the cycleprovided roughly equivalent results of maintaining a constant pH300°C of 6.9, withsignificantly higher radiation fields at lower pH levels. The Guidelines revision


F-3

discussed several chemistry control regimes, including constant pHT of 6.9 forextended burnup, elevated lithium to 3.5 ppm used at the time by Ringhals 2/3/4, andthe constant pHT of 7.4. Caveats were given as to possible fuel corrosion concerns andsteam generator tubing corrosion concerns with operation at lithium above 2.2 ppm. Anote to the figure for the lithium-boron control regimes (Figure 2-4) indicated, "Inpractice, plant-specific graphs should be developed to reflect appropriate coretemperatures." However, guidance was not provided at that time as to which coretemperatures should be used, and algorithms for performing the high temperature pHcalculations were not standardized. The emphasis was to operate at a constant orincreasing pH in the range 6.9 to 7.4, with operation at constant lithium at thebeginning of the cycle until the desired pH was reached, and then coordinating lithiumwith the boron for the remainder of the cycle.

In 1990 PWR Primary Water Chemistry Guidelines: Revision 2 was issued (5). Thisrevision reiterated the results of CRUDSIM predictions of radiation fields based oncrud behavior at pH300°C. However, the figure for various pH regimes (Figure 3-2)specified, "Plant-specific regimes should be adopted using core average temperatures(Tave) . . . ." Lithium levels were given in plots for the various control regimes at 285°C,300°C, and for 315°C. Appendix A of that Guidelines revision presented lithium levelsto provide pH in incremental 0.10 pH unit between 6.70 and 7.50 for 275°C, 280°C,285°C, 295°C, 300°C, 305°C, 310°C, and 315°C. Polynomial expressions for equilibriumconstants as a function of temperature were also provided. These expressions were notfrom the original literature sources, but were fitted expressions. For example, theexpression for the ion product of water was based on an independent fit to Marshalland Franck's data (6) at saturation temperature rather than using the Marshall andFranck correlation; this is the reason values from the polynomial expression for Kw inRevision 2 of the Guidelines are not the same as Marshall and Franck's values.

pH optimization was refined in PWR Primary Water Chemistry Guidelines: Revision 3 (7),and plant-specific regimes using core average temperatures were again specified forcontrol purposes. The perception starting in Revision 1 of the Guidelines was thatoperating at a defined pH at a plant-specific temperature would provide desiredradiation field control, particularly if pHTave of 7.4 could be achieved.

3.0 Calculation of High Temperature pH

When ionic equilibria for high temperature dissociation/hydrolysis reactions becameavailable (mostly from work by Sweeton, Mesmer, and Baes at Oak Ridge at up to300°C), common practice was to calculate the pH at temperature. As differentalgorithms were developed for high temperature pH calculations, a variety of practicesevolved. Ionic equilibria for dissociation/hydrolysis reactions are a function oftemperature, pressure, and ionic strength, and various algorithms used to calculate thehigh temperature pH did not always consider all of these effects. The Marshall-Franck


F-4

algorithm for the ion product of water published in 1981 (6) enabled calculations to beextended to 1000°C and 10,000 bars pressure. Unfortunately, this algorithm used thelog10 of the water density to provide the pressure effect on the ion product of water, andcommon practice was to use saturation pressure derived from a separate algorithm(including PWR Primary Water Chemistry Guidelines: Revision 2), not recognizing thatthis was a pressure effect rather than a density correction (the ion product of waterincreases with pressure) as shown in Table F-1. Although the density difference atsaturation conditions may appear trivial, the effect on Kw is significant (e.g., at 300°C, Kw

at saturation density is 3.927 x 10-12, in molal units, and it increases to 5.242 x 10-12 at 2235

psig pressure, which has a significant impact on the amount of lithium needed toachieve a particular pH).

SYSCHEM and many other codes, including MULTEQ, perform the high temperaturepH calculation at saturation pressure rather than at operating pressure. EPRIstandardized high temperature pH calculations for the reactor coolant in the PWRPrimary Water Chemistry Guidelines: Revision 2 (5). pH calculations in the Guidelinesrevision were based on Tave, and algorithms using derived ionic equilibria (rather thanmore complex algorithms in original sources) were given. The pH Calculator inchemWORKSTM performs pH calculations at temperature using MULTEQ, which hasbeen modified to provide the high temperature pH calculations as obtained withSYSCHEM. Thus, the industry can calculate pH on a consistent basis, which issufficient for chemistry control purposes.

4.0 Composition Of High Temperature Basic Solution

Various pH control programs in the United States focus on operating at a pH at Tave.However, the [H+] in solution at operating conditions is a minor constituent of the ionsin solution, and the pH is highly dependent on the ion product of water, which ishighly dependent on temperature and somewhat less dependent on pressure and ionicstrength. This is illustrated in Figure F-1 for a neutral solution at zero ionic strengthusing the Marshall-Franck correlation for Kw.


F-5

Table F-1The Ion Product of Water Based on the Marshall-Franck Correlation(1) and PWR Primary Water Chemistry Guidelines: Revision2 (EPRI NP-7077)(2) for 0 Ionic Strength

Density (3), g/cm3 Marshall-Franck, Saturation EPRI NP-7077 Marshall-Franck, 2235 psig

Temp., C Saturation 2235 psig log Kw, pHN @ 0 IS log Kw, pHN @ 0 IS log Kw, pHN @ 0 IS0.01 0.99978 1.00754 -14.9375 7.469 -14.9374 7.469 -14.8676 7.434

25 0.99702 1.00389 -13.9948 6.997 -13.9997 7.000 -13.9371 6.96950 0.98799 0.99464 -13.2751 6.638 -13.2709 6.635 -13.2219 6.61175 0.97484 0.98159 -12.7107 6.355 -12.7054 6.353 -12.6586 6.329

100 0.95839 0.96548 -12.2640 6.132 -12.2652 6.133 -12.2105 6.105125 0.93907 0.94668 -11.9111 5.956 -11.9209 5.960 -11.8544 5.927150 0.91706 0.92536 -11.6364 5.818 -11.6508 5.825 -11.5748 5.787175 0.89234 0.90151 -11.4304 5.715 -11.4419 5.721 -11.3621 5.681200 0.86474 0.87496 -11.2878 5.644 -11.2889 5.644 -11.2107 5.605225 0.83387 0.84526 -11.2071 5.604 -11.1951 5.598 -11.1195 5.560250 0.79907 0.81176 -11.1917 5.596 -11.1717 5.586 -11.0912 5.546260 0.78383 0.79701 -11.2055 5.603 -11.1861 5.593 -11.0997 5.550270 0.76768 0.78136 -11.2323 5.616 -11.2165 5.608 -11.1208 5.560280 0.75052 0.76457 -11.2732 5.637 -11.2646 5.632 -11.1566 5.578290 0.73216 0.74652 -11.3303 5.665 -11.3325 5.666 -11.2087 5.604300 0.71241 0.72687 -11.4059 5.703 -11.4224 5.711 -11.2805 5.640310 0.69095 0.70508 -11.5038 5.752 -11.5368 5.768 -11.3780 5.689320 0.66736 0.68058 -11.6295 5.815 -11.6784 5.839 -11.5080 5.754325 0.65458 0.66684 -11.7052 5.853 -11.7602 5.880 -11.5905 5.795330 0.64100 0.65184 -11.7914 5.896 -11.8499 5.925 -11.6878 5.844

1.

William L. Marshall and E. U. Franck, “Ion Product of Water Substance, 0-000oC, 1-10,000 Bars: New International Formulation and its

Background.” J. Phys. Chem. Ref. Data, Vol. 10, No. 2, 1981, pp. 295-304 [Kw is based on temperature and pressure, which is based on density].2. PWR Primary Water Chemistry Guidelines: Revision 2. Research Project 2493. Palo Alto, Calif.: Electric Power Research Institute, November1990. NP-7077 [Kw is based on temperature at saturation pressure].3. Lester Haar, John S. Gallagher, and George S. Kell. NBS/NRC Steam Tables. Washington, D.C.: Hemisphere Publishing Corp. 1984.


F-6

5.4

5.6

5.8

6.0

6.2

6.4

6.6

6.8

7.0

7.2

7.4

7.6

0 25 50 75 100 125 150 175 200 225 250 275 300 325

Temperature, Degrees C

Neu

tral

pH

pH @ Sat. Pressure pH @ P = 2235 psig

Core Temp. at 293.4, 309.2, and 325.0C

Th

Basic

AcidicTc

Tave

Figure F-1Neutral pH at Different Core Temperatures

To illustrate the relative magnitude of ionic concentrations for a typical PWR situation,MULTEQ was used to calculate the lithium necessary for a pHT of 7.20 at several coretemperatures. (Note that MULTEQ uses natural lithium rather than lithium-7). Asshown below, the solution composition at pHT = 7.20 is different for two solutions with1200 and 200 ppm boron and also different at Tc, Tave, and Th.

Temp., Boron, NLi, Molal ConcentrationCore °C ppm ppm [H+] [OH-] [B(OH)4

-] [B2(OH)7

-] [B3(OH)10

-]Tc 293.4 1200 5.230 6.29E-08 8.17E-05 5.83E-04 8.19E-05 8.41E-06Tave 309.2 1200 3.383 6.29E-08 5.78E-05 3.75E-04 5.11E-05 4.38E-06Th 325.0 1200 1.881 6.29E-08 3.46E-05 2.08E-04 2.72E-05 1.96E-06Tc 293.4 200 1.166 6.29E-08 7.58E-05 9.04E-05 2.12E-06 3.63E-08Tave 309.2 200 0.791 6.29E-08 5.41E-05 5.87E-05 1.33E-06 1.91E-08Th 325.0 200 0.459 6.29E-08 3.27E-05 3.28E-05 7.18E-07 8.59E-09

These calculations indicate two important aspects of the solution chemistry foroperating conditions:

• The hydronium ion [H+] is insignificant relative to the concentration of lithium,hydroxyl ion, and major borate species.


F-7

• The composition of the solutions at the same pHT is different at different timesduring the cycle and at different core locations.

For a given lithium and boron concentration, the higher the temperature the higher thepH, as clearly shown in Figure F-2 for a PWR fuel cycle at Tc, Tave, and Th.

In Europe, the pHT control program is based on a common reference temperature of300°C regardless of the plant operating temperature. This has the same effect as usingpOH or pH above neutral as the control parameter:

pOH = pKw - pH

= 2pHN – pH

and

(pH > pHN) = pH - pHN

pH, pHN, and pOH are shown in Figure F-3 as a function of core temperature atpH300°C = 7.2 BOL. As evident in Figure F-3, pH and pHN are very sensitive totemperature, whereas pOH is relatively insensitive. Figure F-4 is a plot of pH atdifferent temperatures using the lithium necessary to obtain pH300°C = 7.20. As evident inthis plot, the higher the temperature, the higher the pH at that temperature. Figure F-5shows pOH for the same lithium levels necessary to provide pH300°C = 7.20. pOH isrelatively insensitive to temperature for a particular lithium-to-boron ratio. One item tobe noted in this regard is that the solubility of transition metal oxides is a function oftemperature and the hydroxide ion concentration (transition metals in solution atoperating conditions exist as hydroxyl complexes). The behavior with respect to crudprecipitation and dissolution behavior can be represented by a series of reactionsinvolving the hydroxyl ion:

M aq+ + + (b)OH

- ↔ M(OH)b

(2 b)− +

M(OH)2 ↔ MO(s) + H2O

Thus, transition metal oxide stability can be examined in terms of pOH and T in orderto separate the effect of local chemistry environment (ion concentrations) from the effectof a change in temperature.


F-8

TEMPERATURES FOR A FUEL CYCLE

6.20

6.30

6.40

6.50

6.60

6.70

6.80

6.90

7.00

7.10

7.20

7.30

7.40

7.50

7.60-1

0 10 30 50 70 90 110

130

150

170

190

210

230

250

270

290

310

330

350

370

390

410

430

450

470

490

510

530

550

570

590

610

630

650

670

690

Time from Initial Power Production, Days

pH a

t Tem

pera

ture

(C

alcu

late

d)

0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

Bor

on, p

pm; L

ithi

um, p

pb

pH @ Th pH @Tave pH @Tc

Boron, ppm Li, ppb

pH @ Tc

pH @ Tave

pH @ Th

Figure F-2Calculated pH at Different Core Temperatures for a Fuel Cycle

6.0

6.2

6.4

6.6

6.8

7.0

7.2

7.4

7.6

7.8

8.0

290 295 300 305 310 315 320 325

Temperature, Deg. F

pH @

Tem

p.

4.0

4.2

4.4

4.6

4.8

5.0

5.2

5.4

5.6

5.8

6.0

pH

n and

pO

H @

Tem

p.

pHt @ BOL pHn @ BOL pOHt @ BOL

Calculations performedchemWORKS .

Core Inlet Core Average Core Outlet

Figure F-3Illustration of pH, pHN, and pOH as a Function of Temperature for pH = 7.20 at BOL


F-9

p 300

7.00

7.05

7.10

7.15

7.20

7.25

7.30

7.35

7.40

7.45

7.500

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

Boron, ppm

pH a

t Tem

pera

ture

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

Lith

ium

, ppm

pHt=320 pHt=315 pHt=310 pHt=305 pHt=300

pHt=295 pHt=290 pHt=285 Li, ppm

Figure F-4pH at Various Temperatures Using Lithium for pH300°C = 7.20

p

4.10

4.15

4.20

4.25

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

Boron, ppm

pO

H a

t Tem

per

atu

re

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

Lith

ium

, ppm

pOHt=285 pOHt=290 pOHt=295pOHt=300 pOHt=305 pOHt=310pOHt=315 pHt=320 Li, ppm

Figure F-5pOH at Various Temperatures Using Lithium for pH300°C = 7.20


F-10

5.0 CRUDSIM Calculations Based on pH

CRUDSIM is an EPRI code that calculates the distribution of iron and cobalt activitiesfrom nickel ferrite in a PWR cycle based on the boron and lithium. Power level andtemperature are considered in the code. CRUDSIM is overly simplistic in that the cruddistribution is based only on nickel ferrite solubility, equilibrium throughout the RCS,and the absence of boiling. Solubility of different oxide forms, colloid behavior, crudredistribution during cooldowns, etc. are not considered in this model. The focus ofCRUDSIM is on iron behavior, particular deposition on the core. However, even withthese limitations, inferences can be obtained from the code results.

The crud distribution based on the following pH control programs was estimated withCRUDSIM:

• Modified Chemistry Control of pH 6.9 to 7.4 based on Tave (309°C), 2.2 ppmmaximum lithium

• Modified Chemistry Control of pH 6.9 to 7.4 based on T300°C, 2.2 ppm maximumlithium

• Coordinated Chemistry Control for pH 6.9 based on T300°C




The results of CRUDSIM calculations presented here are not intended to defineoptimum pH operating regime based on Tave or 300°C, or pOH, but are presented onlyto illustrate the sensitivity of crud deposition using CRUDSIM based on differentcontrol regimes. The boron concentration for a PWR with an extended fuel cycleshown in Figure F-6 was the basis of the CRUDSIM calculations. The plant operationwas assumed to be for 100% power for the entire cycle. Note in Figure F-6 that thecommon practice of basing the pH control program on Tave (309°C) rather than 300°C orpOH results in less lithium to obtain a particular pH at temperature (e.g., 1730 ppmboron and 2.33 ppm lithium at Tave = 309°C gives a pH of 6.90, whereas at 300°C the pHis only 6.78). Thus, plants basing lithium and pH on Tave may have BOL pH levels lessthan 6.90 relative to 300

oC.

Figures F-7 through F-13 show the results of the CRUDSIM calculations for thedifferent pH control programs. These figures suggest the following:


F-11

• Higher lithium levels (e.g., lower pOH or higher pH300°C) results in less irondeposition on the core (Figure F-7) and less cobalt activity in the loops (Figures F-10and F-12).

• Modified Chemistry Control based on Tave results in increased iron deposition andcobalt activity on the core relative to Modified Chemistry Control based on 300°C orpOH (Figures F-7, F-9, and F-11).

• Stable pH or pOH will result in less cobalt activity in the loops than a variablecontrol program such as Modified Chemistry Control (Figure F-10 and F-12) with amaximum lithium level of 2.2 ppm.

• Low pH followed by increasing pH may release activity from the core (Figure F-13).

• The 6.9 pH300°C results in less loop cobalt activity than the Modified ChemistryControl based on either Tave or 300°C with a maximum lithium level of 2.2 ppm.However, the inventory of activity on the core is the highest of the cases studied.

The following summarized the CRUDSIM calculations for the cases studied:

Control Cobalt-58, Ci Cobalt-60,Ci Core Iron,Program Basis pHT Core Loop Total Core Loop Total kg

Modified,2.2

Tave 6.9-7.4 2,717 3,006 5,723 193 368 561 8.30

ppm max. Li 300oC 6.9-7.4 2,096 2,997 5,092 149 362 511 6.93Coordinated 300oC 6.9 4,652 2,802 7,454 313 320 633 12.38

7.0 3,159 2,571 5,730 201 283 485 9.587.1 1,812 2,175 3,987 107 226 333 6.777.2 632 1,431 2,062 33 133 166 3.67


F-12

C CU ON N C U S N U

0

200

400

600

800

1000

1200

1400

1600

1800

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Bor

on, p

pm

0

1000

2000

3000

4000

5000

6000

7000

Lith

ium

, ppb

B (ppm)

pH 7.2/pOH 4.17 (300C)

pH 7.1/pOH 4.28 (300C)

pH 7.0/pOH 4.38 (300C)

pH 6.9/pOH 4.49 (300C)

Mod. pH 6.9 - 7.4 (300C)

Mod. pH 6.9 - 7.4 (309C)

Figure F-6Boron and Lithium Used for pH Calculation and CRUDSIM Input


F-13

0

2

4

6

8

10

12

14

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Cor

e Ir

on, k

g

pH 6.9/pOH 4.49 (300C)

Mod. pH 6.9 - 7.4 (309C)

Mod. pH 6.9 - 7.4 (300C)

pH 7.0/pOH 4.38 (300C)

pH 7.1/pOH 4.28 (300C)

pH 7.2/pOH 4.17 (300C)

Figure F-7CRUDSIM Crud Distribution for Different pH Controls: Core Iron

15

20

25

30

35

40

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Loop

Iro

n, k

g

pH 7.2/pOH 4.17 (300C)

pH 7.1/pOH 4.28 (300C)

pH 7.0/pOH 4.38 (300C)

Mod. pH 6.9 - 7.4 (300C)

Mod. pH 6.9 - 7.4 (309C)

pH 6.9/pOH 4.49 (300C)

Figure F-8CRUDSIM Crud Distribution for Different pH Controls: Loop Iron


F-14

p

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Cor

e 58

Co,

Ci

pH 6.9/pOH 4.49 (300C)

Mod. pH 6.9 - 7.4 (309C)

Mod. pH 6.9 - 7.4 (300C)

pH 7.0/pOH 4.38 (300C)

pH 7.1/pOH 4.28 (300C)

pH 7.2/pOH 4.17 (300C)

Figure F-9CRUDSIM Crud Distribution for Different pH Controls: Core Cobalt-58

0

500

1,000

1,500

2,000

2,500

3,000

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Loop

58 C

o, C

i

Mod. pH 6.9 - 7.4 (309C)

Mod. pH 6.9 - 7.4 (300C)

pH 6.9/pOH 4.49 (300C)

pH 7.0/pOH 4.38 (300C)

pH 7.1/pOH 4.28 (300C)

pH 7.2/pOH 4.17 (300C)

Figure F-10CRUDSIM Crud Distribution for Different pH Controls: Loop Cobalt-58


F-15

p

0

50

100

150

200

250

300

350

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Cor

e 60

Co,

Ci

pH 6.9/pOH 4.49 (300C)

Mod. pH 6.9 - 7.4 (309C)

Mod. pH 6.9 - 7.4 (300C)

pH 7.0/pOH 4.38 (300C)

pH 7.1/pOH 4.28 (300C)

pH 7.2/pOH 4.17 (300C)

Figure F-11CRUDSIM Crud Distribution for Different pH Controls: Core Cobalt-60

0

50

100

150

200

250

300

350

400

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Loop

60C

o, C

i

Mod. pH 6.9 - 7.4 (309C)

Mod. pH 6.9 - 7.4 (300C)

pH 6.9/pOH 4.49 (300C)

pH 7.0/pOH 4.38 (300C)

pH 7.1/pOH 4.28 (300C)

pH 7.2/pOH 4.17 (300C)

Figure F-12CRUDSIM Crud Dirstibution for Different pH Controls: Loop Cobalt-60


F-16

p

-4

-2

0

2

4

6

8

10

12

14

16

18

0 25 50 75 100

125

150

175

200

225

250

275

300

325

350

375

400

425

450

475

500

525

Days in Cycle

Cor

e Ir

on S

olul

bili

ty, %

Dep

osit

ion

Mod. pH 6.9 - 7.4 (309C)pH 6.9/pOH 4.49 (300C)Mod. pH 6.9 - 7.4 (300C)pH 7.0/pOH 4.38 (300C)pH 7.1/pOH 4.28 (300C)pH 7.2/pOH 4.17 (300C)

Figure F-13CRUDSIM Crud Distribution for Different pH Controls: Core Iron Solubility

5.1 Inferences from CRUDSIM Calculations for Li-B programs

Although this evaluation is not intended to define the optimum pH control regime,high, constant pH (e.g., a coordinated program) seems to be indicated. A coordinated7.3 or 7.4 based on pH300°C most likely is not realistic based on lithium limitationspresently imposed by fuel vendors. Coordinated chemistry programs at elevated pHare consistent with pH control principles in Volume 1 of these Guidelines in that PWRswith concern for early cycle crud deposition onto the core should consider a constantpH based on Tave as high as possible up to pH 7.4, consistent with the integrated lithiumexposure permitted by the fuel vendor.

One should note that crud deposition from boiling is expected to be more a function ofthe mass evaporation rate than solubility driven deposition controlled by pH. The pHcontrol program, particularly during the early portion of the fuel cycle, can limit releasefrom plant surfaces by reducing corrosion, and thus may affect deposition by reducingthe source term rather than by affecting deposition/solubility on the fuel.


F-17

6.0 Practical Considerations

The practice in the United States of basing reactor coolant chemistry control on lithiumto provide a particular pH at Tave results in a lower operating pH at plants withTave>300°C, at the beginning of a fuel cycle, particularly for high duty cores which haverelatively high boron at BOL. For example, a plant with Tave = 310°C and boron = 1500ppm at BOL will operate with only 1.90 ppm lithium to achieve pH310°C = 6.90, whereas2.66 ppm lithium is required to provide pH300°C = 6.90. Figure F-14 illustrates the effectof basing the lithium control on pHTave rather than pH300°C for Modified ChemistryControl. Less lithium is required to achieve a particular pH at Tave>300°C using theconvention based on Tave. As evident in Figure F-14, basing chemistry control on Tave

requires significantly less lithium at the beginning of the fuel cycle relative to usingpH300°C, and the control strategy at the end of the cycle is also different.

Another area where basing pH on 300°C versus average core temperature producespractical differences in operation is during power changes. Most PWRs adjust thelithium for changes in power level resulting in changes in Tave, including powercoastdowns. By using pH300°C rather than pHTave, such adjustments during short powertransients would not be necessary.

If changes are to be made in U.S. plants in order to be consistent with Europeanpractices for reactor coolant chemistry control, NSSS vendors and fuel vendors mustreview the potential impact of changing the pH control program.


F-18

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

Boron, ppm

Lith

ium

, ppm

6.80

6.90

7.00

7.10

7.20

7.30

7.40

7.50

pH a

t Tem

pera

ture

Li - Hi Limit, 300C Li - Target, 300C Li - Low Limit, 300C

Li - Hi Limit, 310C Li - Target, 310C Li - Low Limit, 310C

pHt, 310C pHt, 300C

Figure F-14Modified Chemistry Control Based on Tave = 310°C and T = 300° C

7.0 References

1. The Role of Coolant Chemistry in PWR Radiation-Field Buildup. Research Project825-2. Palo Alto, Calif.: Electric Power Research Institute, October 1985. NP-4247.

2. The CORA-II Model of PWR Corrosion-Product Transport. Research Project 825-2Palo Alto, Calif.: Electric Power Research Institute, September 1985. NP-4246.

3. PWR Primary Water Chemistry Guidelines. Special Report. Palo Alto, Calif.:Electric Power Research Institute, September 1986. NP-4762-SR.

4. PWR Primary Water Chemistry Guidelines: Revision 1. Special Report. Palo AltoCalif.: Electric Power Research Institute, August 1988. NP-5960-SR.

5. PWR Primary Water Chemistry Guidelines: Revision 2. Research Project 2493. PaloAlto, Calif.: Electric Power Research Institute, November 1990. NP-7077.

6. William L. Marshall and E. U. Franck, "Ion Product of Water Substance, 0-1000oC, 1-

10,000 Bars: New International Formulation and its Background." J. Phys. Chem.Ref. Data, Vol. 10, No. 2, 1981, pp. 295-304.


F-19

7. PWR Primary Water Chemistry Guidelines: Revision 3. Research Project 2493. PaloAlto, Calif.: Electric Power Research Institute, November 1995. TR-105714.

8. In-Pile Loop Studies on the Effect of PWR Coolant pH on Corrosion ProductRadionuclide Deposition. Research Project 22295-4. Palo Alto, Calif.: ElectricPower Research Institute, February 1992. TR-100156.

9. A Survey of the Effect of Primary Coolant pH on Westinghouse PWR PlantRadiation Fields. Research Project 2493-05. Palo Alto, Calif.: Electric PowerResearch Institute, November 1994. TR-104180.

EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA800.313.3774 • 650.855.2121 • [email protected] • www.epri.com

WARNING: This Document containsinformation classified under U.S. ExportControl regulations as restricted fromexport outside the United States. You

are under an obligation to ensure that you have alegal right to obtain access to this informationand to ensure that you obtain an export licenseprior to any re-export of this information. Specialrestrictions apply to access by anyone that is nota United States citizen or a Permanent UnitedStates resident. For further informationregarding your obligations, please see theinformation contained below in the section titled“Export Control Restrictions.”

© 1999 Electric Power Research Institute (EPRI), Inc.All rights reserved. Electric Power ResearchInstitute 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

Program: TR-105714-V1R4

Nuclear Power

About EPRI

EPRI creates science and technology solutions forthe global energy and energy services industry.U.S. electric utilities established the Electric PowerResearch Institute in 1973 as a nonprofit researchconsortium for the benefit of utility members, theircustomers, and society. Now known simply as EPRI,the company provides a wide range of innovativeproducts and services to more than 1000 energy-related organizations in 40 countries. EPRI’smultidisciplinary team of scientists and engineersdraws on a worldwide network of technical andbusiness expertise to help solve today’s toughestenergy and environmental problems.

EPRI. Electrify the World

Export Control RestrictionsAccess to and use of EPRI Intellectual Property is grantedwith the specific understanding and requirement thatresponsibility for ensuring full compliance with all applicableU.S. and foreign export laws and regulations is beingundertaken by you and your company. This includes anobligation to ensure that any individual receiving accesshereunder who is not a U.S. citizen or permanent U.S.resident is permitted access under applicable U.S. and foreignexport laws and regulations. In the event you are uncertain whether you oryour company may lawfully obtain access to this EPRIIntellectual Property, you acknowledge that it is your obligation to consult with your company’s legal counsel todetermine whether this access is lawful. Although EPRI maymake available on a case by case basis an informal assessmentof the applicable U.S. export classification for specific EPRIIntellectual Property, you and your company acknowledgethat this assessment is solely for informational purposes andnot for reliance purposes. You and your company acknowledge that it is still the obligation of you and yourcompany to make your own assessment of the applicableU.S. export classification and ensure compliance accordingly.You and your company understand and acknowledge yourobligations to make a prompt report to EPRI and the appropriate authorities regarding any access to or use ofEPRI Intellectual Property hereunder that may be in violationof applicable U.S. or foreign export laws or regulations.

PWR Primary Water Chemistry Guidelines - Miroslav · PDF fileEPRI • 3412 Hillview Avenue,...

Documents

Transcript of PWR Primary Water Chemistry Guidelines - Miroslav · PDF fileEPRI • 3412 Hillview Avenue,...