ML18214A132

NUREG-2224 Dfc, Dry Storage and Transportation of High Burnup Spent Nuclear Fuel - Draft Report for Comment

ML18214A132
Person / Time
Issue date:	07/31/2018
From:	Reed W A Office of Nuclear Material Safety and Safeguards
To:
Meyd, Donald
References
NUREG-2224 DFC
Download: ML18214A132 (129)

Text

NUREG-2224 Dry Storage and Transportation of High Burnup Spent Nuclear Fuel Draft Report for Comment Offic e of N u clear Material Safety a n d Safeg u ards NRC Reference Material As of November 1999, you may electronically access NUREG-series publications and other NRC records at the http://www.nrc.gov/reading

-rm.html. Publicly released records include, to name a few, NUREG

-series publications; Federal Register notices; applicant, licensee, and vendor documents and correspondence; NRC correspondence and internal memoranda; bulletins and information notices; inspection and investigative reports; licensee event reports; and Commission papers and their attachments.

NRC publications in the NUREG series, NRC regulations, and Title 10, Energy ,Code of Federal Regulations may also be purchased from one of these two sources. 1. The Superintendent of DocumentsU.S. Government Publishing OfficeWashington, DC 20402-0001Internet: http://bookstore.gpo.govTelephone:

1-866-512-1800Fax: (202) 512-2104 2. The National Technical Information Service5301 Shawnee RoadAlexandria, VA 22161-0002http://www.ntis.gov 1-800-553-6847 or, locally, (703) 605-6000A single copy of each NRC draft report for comment is available free, to the extent of supply, upon written request as follows:

U.S. Nuclear Regulatory Commission Office of Administration Multimedia, Graphics and Storage & Distribution Branch Washington, DC 20555-0001 E-mail: distribution.resource@nrc.govFacsimile:

(301) 415-2289Some publications in the NUREG series that are posted at the http://www.nrc.gov/reading

-rm/doc-collections/nuregs are updated periodically and may differ from the last printed version. Although references to material found on a Web site bear the date the material was accessed, the material available on the date cited may subsequently be removed from the site.

N on-NRC Reference Material Documents available from public and special technical libraries include all open literature items, such as books, journal articles, transactions, Federal Register notices, Federal and State legislation, and congressional reports. Such documents as theses, dissertations, foreign reports and translations, and non

-NRC conference proceedings may be purchased from their sponsoring organization.

Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at The NRC Technical Library Two White Flint North 11545 Rockville Pike Rockville, MD 20852-2738 These standards are available in the library for reference use by the public.

Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from American National Standards Institute 11 West 42 nd Street New York, NY 10036

-8002 http://www.ansi.org (212)642-4900 AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATION S Legally binding regulatory requirements are stated only in laws; NRC regulations; licenses, including technical specifications; or orders, not in NUREG

-series publications. The views expressed in contractor

-prepared publications in this series are not necessarily those of the NRC.

The NUREG series comprises (1) technical and administrative reports and books prepared by the staff (NUREG-XXXX) or agency contractors (NUREG/CR-XXXX), (2) proceedings of conferences (NUREG/CP-XXXX), (3) reports resulting from international agreements (NUREG/IA-XXXX), (4) brochures (NUREG/BR

-XXXX), and (5) compilations of legal decisions and orders of the Commission and (NUREG-0750). DISCLAIMER:

This report was prepared as an account of work sponsored by an agency of the U.S. Government. Neither the U.S. Government nor any agency thereof, nor any employee, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any information, apparatus, product, or process disclosed in this publication, or represents that its use by such third party would not infringe privately owned rights.

SR-CR 10/2017 Dry Storage and Transportati on of High Burnup

Spent N uclear Fuel Draft R eport f or Comment M a nu script Completed:

April 2018 Dat e P u blis h ed: J u ly 2018 Prepared b y: T.A hn H.A kh a v a nn i k G.Bjorkman F.C. C h a ng W.Reed A.Rigato D.T a ng R.D. Torres B.H. White

V.W i lson O f f i c e o f N u c l e ar Mat e rialSa f e ty and Sa f eg uar d s NUREG-2224 C O M M E N T S O N D RA F T R E P O RT Any interest ed party may submit comments on this report f or considerati on by t he staff of t he U.S. Nuclear Regulatory Commission (NRC).

Comments m ay be accompanied by additional relevant informati on or supporti ng data. Please specify t he report num ber NUREG-222 4 in your comments, and send them by the end of the comment peri od specified in t he Federal Register notic e announci ng t he availability of this report. Addr e ss e s: Y ou may submit comments by any one of t he followi ng methods. Please include Docket I D NRC-2 018-0 066 i n the subject l i ne of y our comments. Comments submitt ed in writing or in electronic form will be posted on the NRC W eb site and on t he Federal rulemaking Web si t e (http://www.regulations.gov

). Federa l R ule mak ing Web site: G o t o http://www.regulations.gov and searc h for documents fil ed under D ocket I D NRC-2 018-0 066. Address ques tions about NRC doc kets to J ennifer Borges; t elephone: 301-2 87-9 127; e-m ail: Jennifer.Borges@nrc.govFor any questions about the material i n this r eport, pl ease contac t: W endy R eed, 301-4 15-7 213 or, e-m ail at W endy.Reed@nrc.gov. Please be aware that any comments that y ou submit to t he NRC w ill be consider ed a public record and entere d int o the Agencywi de Documents Access and M anagement System. D o not provi de informati on y ou woul d not want t o be publicly available.

ABSTRACT 1 Time-dependent changes on the cladding performance of high burnup (HBU) spent nuclear fuel 2 (SNF) are all primarily driven by the fuel's temperature, rod internal pressure (and 3 corresponding pressure

-induced cladding hoop stresses), and the environment during dry 4 storage or transport operations. Historically, the potential for these changes to compromise the 5 analyzed fuel configuration in dry storage systems and transportation packages has been 6 addressed through safety review guidance. This guidance defines adequate fuel conditions, 7 including peak cladding temperatures during short

-term loading operations to prevent or 8 mitigate degradation of the cladding. The purpose of this report is to expand the technical basis 9 in support of that guidance, as it pertains to the mechanism of hydride reorientation in HBU SNF 10 cladding. 11 Hydride reorientation is a process in which the orientation of hydrides precipitated in HBU SNF 12 cladding during reactor operation change s from the circumferential

-axial to the radial

-axial 13 direction. Research results over the last decade have shown that hydride reorientation can still 14 occur at temperatures and stresses lower than those assumed in the current staff review 15 guidance. Therefore, the U.S. Nuclear Regulatory Commission (NRC) has since sponsored 16 additional research to better understand whether hydride reorientation could affect the 17 mechanical behavior of HBU SNF cladding and compromise the fuel configuration analyzed in 18 dry storage systems and transportation packages.

19 This report provides an engineering assessment of the results of research on the mechanical 20 performance of HBU SNF following hydride reorientation. Based on the conclusions of that 21 assessment, the report then presents example approaches for licensing and certification of HBU 22 SNF for dry storage (under Title 10 of the Code of Federal Regulations (10 CFR) Part 72 , 23 "Licensing Requirements for the Independent Storage of Spent Nuclear Fuel and High

-Level 24 Radioactive Waste, and Reactor

-Related Greater Than Class C Waste") and transportation 25 (under 10 CFR Part 71 , "Packaging and Transportation of Radioactive Material"

). 26 The information in this report is not intended for use in applications for wet storage facilities or 27 monitored retrievable storage installations licensed under 10 CFR Part 72.

28 Nothing contained in this report is to be construed as having the force or effect of regulations.

29 Comments regarding errors or omissions, as well as suggestions for improvement of this 30 NUREG should be sent to the Director, Division of Spent Fuel Management, U.S. Nuclear 31 Regulatory Commission, Washington, D.C., 20555

-0001. 32 Paperwork Reduction Act 33 34 This NUREG provides guidance for implementing the mandatory information collections in 10 35 CFR Parts 71 and 72 that are subject to the Paperwork Reduction Act of 1995 (44 U.S.C. 3501 36 et. seq.). These information collections were approved by the Office of Management and Budget 37 (OMB) under control numbers 3150

-0008 and 3150

-0132. Send comments regarding this 38 information collection to the Information Services Branch, U.S. Nuclear Regulatory Commission, 39 Washington, DC 20555

-0001, or by e

-mail to Infocollects.Resource@nrc.gov, and to the Desk 40 Officer, Office of Information and Regulatory Affairs, NEOB

-10202, (3150

-0008, 3150

-0132) 41 Office of Management and Budget, Washington, DC 20503.

42 43 Public Protection Notification 1 The NRC may not conduct or sponsor, and a person is not required to respond to, a collection 2 of information unless the document requesting or requiring the collection displays a currently 3 valid OMB control number.

4 v CONTENTS 1 ABSTRACT ...................................................................................................................

iii 2 CONTENTS .................................................................................................................... v 3 LIST OF FIGURES

........................................................................................................

vii 4 LIST OF TABL ES. .........................................................................................................

ix 5 ACKNOWLEDGMENTS

................................................................................................

xi 6 ABBREVIATIONS AND ACRONYMS .........................................................................

xiii 7 1 INTRODUCTION

...................................... 8 1.1 Back ground ...........................................................

9 1.2 Fuel Cladding Performance and Staff's Review Guidance

................................1-2 10 1.3 Cladding Creep .................................................................................................1-4 11 1.4 Effects of Hydrogen on Cladding Mechanical Performance

...............................1-5 12 1.5 Hydride Reorientation

.......................................................................................1-7 13 1.5.1 Hydride Dissolution and Precipitation

....................................................1-8 14 1.5.2 Fuel Cladding Fabrication Process

...................................................... 1-10 15 1.5.3 End-Of-Life Rod Internal Pressures and Cladding Hoop 16 Stresses .............................................................................................. 1-11 17 1.5.3.1 End-Of-Life Rod Internal Pressures for Pressurized

-18 Water Reactor Fuel Rods

..................................................... 1-11 19 1.5.3.2 Gas Temperatures for Fuel Rods During Drying

-20 Transfer, Storage and Transportation

................................... 1-14 21 1.5.3.3 Peak Cladding Hoop Stresses for Pressurized

-Water 22 Reactor Fuel Rods During Drying

-Transfer and 23 Storage/Transport Operations

.............................................. 1-15 24 1.5.4 Ring Compression Testing

.................................................................. 1-18 25 1.5.5 Staff's Assessment of Ring Compression Testing Results

................... 1-24 26 2 ASSESSMENT OF STATIC BENDING AND FATIGUE STRENGTH 27 RESULTS ON HIGH BURNUP SPENT NUCLEAR FUE L................................ 2-1 28 2.1 Introduction

.......................................................................................................2-1 29 2.2 Cyclic Integrated Reversible Fatigue Tester

......................................................2-1 30 2.3 Application Of The Static Test Results

..............................................................2-4 31 2.3.1 Spent Fuel Rod Behavior in Bending. ....................................................2-5 32 2.3.2 Composite Behavior o f a Spent Fuel Rod. .............................................2-6 33 2.3.3 Calculation of Cladding Strain Using Factored Cladding-Only 34 Properties ............................................................................................ 2-10 35 2.3.3.1 Two Alternatives for Calculating Cladding Stress a nd 36 Strain During Drop Accidents

................................................ 2-13 37 2.3.4 Applicability to Dry Storage and Transportation

................................... 2-14 38 2.3.4.1 Use of Static Test Results to Evaluate Safety Margins 39 i n a n HAC Side Drop Event

.................................................. 2-17 40 2.3.4.2 Dynamic Response o f a Fuel Rod ........................................ 2-19 41 2.3.4.3 Seismic Response o f a Fuel Rod .......................................... 2-19 42 2.4 Application of Fatigue Test Results

................................................................. 2-20 43 vi 2.4.1 Lower Bound Fatigue S-N Curves ....................................................... 2-20 1 2.4.2 Fatigue Cumulative Damage Model

..................................................... 2-22 2 2.4.3 Applicability to Storage and Transportation

......................................... 2-23 3 3 DRY STORAGE OF HIGH BURNUP SPENT NUCLEAR FUEL ...................... 3-1 4 3.1 Introduction

.......................................................................................................3-1 5 3.2 Uncanned Fuel (Intact and Undamaged Fuel)

...................................................3-4 6 3.2.1 Leaktight Confinement

...........................................................................3-6 7 3.2.2 Non-Leaktight Confinement

...................................................................3-7 8 3.2.3 Dry Storage U p To 20 Years

............................................................... 3-10 9 3.2.4 Dry Storage Beyond 20 Years

............................................................. 3-11 10 3.2.4.1 Supplemental Results from Confirmatory 11 Demonstration

...................................................................... 3-11 12 3.2.4.1.1 Initial Licensing or Certification

........................... 3-12 13 3.2.4.1.2 Renewal Applications

......................................... 3-12 14 3.2.4.2 Supplemental Safety Analyses

............................................. 3-12 15 3.2.4.2.1 Materials and Structural

...................................... 3-13 16 3.2.4.2.2 Confinement

....................................................... 3-13 17 3.2.4.2.3 Thermal .............................................................. 3-14 18 3.2.4.2.4 Criticality

............................................................. 3-15 19 3.2.4.2.5 Shielding ............................................................. 3-17 20 3.3 Canned Fuel (Damaged Fuel) ......................................................................... 3-19 21 4 TRANSPORTATION OF HIGH BURNUP SPENT NUCLEAR FUEL

............... 4-1 22 4.1 Introduction

.......................................................................................................4-1 23 4.2 Uncanned Fuel (Intact and Undamaged Fuel)

...................................................4-4 24 4.2.1 Leaktight Containment

...........................................................................4-7 25 4.2.2 Non-Leaktight Containment

...................................................................4-7 26 4.2.3 Direct Shipment from the Spent Fuel Pool and Shipment o f 27 Previously Dry

-Stored Fuel (U p To 20 Years since Fuel was 28 Initially Loaded)

................................................................................... 4-11 29 4.2.4 Shipment of Previously Dry

-Stored Fuel (Beyond 20 Years 30 since Fuel was Initially Loaded) ........................................................... 4-12 31 4.2.4.1 Supplemental Data from Confirmatory Demonstration

.......... 4-12 32 4.2.4.2 Supplemental Safety Analyses

............................................. 4-12 33 4.2.4.2.1 Materials and structural

...................................... 4-13 34 4.2.4.2.2 Containment

....................................................... 4-13 35 4.2.4.2.3 Thermal .............................................................. 4-14 36 4.2.4.2.4 Criticality

............................................................. 4-15 37 4.2.4.2.5 Shielding ............................................................. 4-18 38 4.3 Canned F uel ................................................................................................... 4-20 39 5 CONCLUSIONS

................................................................................................ 5-1 40 6 REFERENCES

.................................................................................................. 6-1 41 7 GLOSSARY ...................................................................................................... 7-1 42 vii LIST OF FIGURES 1 Figure 1-1 Hydride Content [H] and Distribution (Average , Inner 2/3 Diameter Of 2 Cladding) i n HBU SNF Cladding (From Billone et al., 2013) .............................. 1-6 3 Figure 1-2 Dissolution (C d) and Precipitation (C p) Concentration Curves Based o n t he 4 Data o f Kammenzind e t al. (1996) for Non-Irradiated Zircaloy

-4 (Zry-4) ............ 1-9 5 Figure 1-3 Publicly-Available Data Collected b y EPRI For PWR End-Of-Life Rod 6 Internal Pressures at 25°C................................................................................. 1-12 7 Figure 1-4 Calculated Rod Internal Pressures for the First 10 Cycles o f t he Watts 8 Bar Nuclear Plant Unit 1 Reactor Under Vacuum Drying Conditions

............... 1-13 9 Figure 1-5 Aggregated Measured and Calculated Values for End-Of-Life Rod Internal 10 Pressures for PWR Fuel Rods. Pressures are Presented at 25 °C (77 °F) .... 1-14 11 Figure 1-6 Calculated Rod Internal Pressure a s a Function o f the Spatially Averaged 12 Gas Temperature for PWR Fuel Rods (i.e. Standard Rods), ZIRLO-Clad 13 IFBA Rods With Hollow (Annular) Blanket Pellets, a nd ZIRLO-Clad IFBA 14 Rods with Solid Blanket Pellets

......................................................................... 1-15 15 Figure 1-7 Fuel Cladding Tube with Stress Element

.......................................................... 1-16 16 Figure 1-8 Calculated Values of Cladding Hoop Stress Vs.

the Spatially

-Averaged 17 Internal Gas Temperature for Standard 17x17 PWR Fuel Rods with 18 Corrosion Layers of 10 µm, 40 µm, and 80 µm................................................. 1-17 19 Figure 1-9 PWR Hoop Stress a s a Function of Internal Gas Temperature for 17 x 17 20 IFBA Fuel Rods (for Both Hollow Blanket and Solid Blanket Pellets) with a 21 40-µm Corrosion Layer

...................................................................................... 1-18 22 Figure 1-10 RCT o f a Sectioned Cladding Ring Specimen i n ANL's Instron's 23 8511 Test Setup

................................................................................................. 1-19 24 Figure 1-11 Ductility Vs. RCT for Two PWR Cladding Alloys Following Slow Cooling 25 from 400°C at Peak Target Hoop Stresses of 110 Mpa and 140 Mpa

............. 1-20 26 Figure 1-12 Ductility Data, as Measured b y RCT , for As-Irradiated Zircaloy

-4 a nd 27 Zircaloy-4 Following Cooling from 400 °C Under Decreasing Internal 28 Pressure and Hoop Stress Conditions

.............................................................. 1-21 29 Figure 1-13 Ductility Data, as Measured b y RCT , for As-Irradiated ZIRLO a nd ZIRLO 30 Following Cooling from 400 °C Under Decreasing Internal Pressure a nd 31 Hoop Stress Conditions

..................................................................................... 1-22 32 Figure 1-14 Ductility Data, as Measured By RCT , for As-Irradiated M5 a nd M5 33 Following Cooling from 400 °C Under Decreasing Internal Pressure a nd 34 Hoop Stress Conditions

..................................................................................... 1-23 35 Figure 1-15 Geometric Models for Spent Fuel Assemblies i n Transportation Packages .... 1-25 36 Figure 2-1 Horizontal Layout o f ORNL U-Frame Setup, Rod Specimen and Three 37 Lvdts for Curvature Measurement, and Front View of CIRFT Installed i n 38 ORNL Hot Cell ..................................................................................................... 2-2 39 Figure 2-2 Schematic Diagram of End and Side Drop Accident Scenarios

......................... 2-5 40 Figure 2-3 Typical Composite Construction o f a Bridge ....................................................... 2-6 41 viii Figure 2-4 Influence o f cg Position on Composite Beam Stiffness

...................................... 2-7 1 Figure 2-5 Images of Cladding

-Pellet Structure i n HBU SNF Rod ....................................... 2-8 2 Figure 2-6 Approximate Extreme Fiber Tensile Stresses Between Pellet

-Pellet Crack

...... 2-9 3 Figure 2-7 Comparison of CIRFT Static Bending Results with Calculated PNNL 4 Moment Curvature (Flexural Rigidity) Derived from Cladding-Only 5 Stress-Strain Curve

............................................................................................ 2-10 6 Figure 2-8 Characteristic Points on Moment-Curvature Curve

.......................................... 2-11 7 Figure 2-9 High Magnification Micrograph Showing Radial Hydrides o f a HBR HBU 8 SNF Hydride-Reoriented Specimen Tested Under Phase II (Specimen 9 HR1 Results Shown; Hydrogen Content 360-400 Wpp m) ............................ 2-15 10 Figure 2-10 Representative Conditions Used for Radial Hydride Treatment for 11 Preparation o f HBR HBU SNF Hydride-Reoriented Specimens Tested 12 Under Phase II ................................................................................................... 2-16 13 Figure 2-11 Plots of Half o f the Cladding Strain Range

(/2) a nd the Maximum Strain 14 (//max) a s a Function of Number of Cycles to Failure........................................ 2-21 15 Figure 2-12 CIRFT Dymanic (Fatigue) Test Results for As-Irradiated a nd 16 Hydride-Reoriented H.B. Robinson Zircaloy

-4 HBU Fuel Rods........................ 2-22 17 Figure 3-1 Example Licensing and Certification Approaches for Dry Storage of High 18 Burnup Spent Nuclear Fuel

................................................................................. 3-3 19 Figure 3-2 First Approach for Evaluating Design

-Bases Drop Accidents During 20 Dry Storage

.......................................................................................................... 3-5 21 Figure 3-3 Second Approach for Evaluati on of Design-Bases Drop Accidents During 22 Dry Storage

.......................................................................................................... 3-6 23 Figure 4-1 Example Approaches for Approval of Transportation Packages with High 24 Burnup Spent Nuclear Fuel

................................................................................. 4-3 25 Figure 4-2 First Approach for Evaluation of Drop Accidents During Transport

.................... 4-5 26 Figure 4-3 Second Approach for Evaluation of Drop Accidents During Transport

.............. 4-6 27 Figure 4-4 Evaluation of Vibration Normally Incident to Transport

....................................... 4-7 28 ix LIST OF TABLES 1 Table 2-1 Comparison of Average Flexural Rigidity Results between CIRFT Static 2 Testing a nd PNNL Cladding-Only Data............................................................. 2-12 3 Table 2-2 Characteristic Points and Quantities Based on Moment-Curvature Curves

..... 2-12 4 Table 2-3 Properties of PWR 15 x 15 SNF R od ................................................................ 2-17 5 Table 2-4 Summary of CIRFT Dynamic Test Results for As-Irradiated and Hydride

-6 Reoriented HBR HBU SNF

................................................................................ 2-20 7 Table 2-5 Coordinates for Lower

-Bound Enveloping S

-N Curve for the HBR HBU 8 SNF Rods........................................................................................................... 2-21 9 Table 3-1 Fractions of Radioactive Materials Available for Release from HBU SNF 10 Under Conditions o f Dry Storage

......................................................................... 3-8 11 Table 4-1 Fractions of Radioactive Materials Available for Release from HBU SNF 12 Under Conditions Of Transport

............................................................................ 4-9 13

xi ACKNOWLEDGMENTS 1 The working group is very grateful to M. Billone (Argonne National Laboratory) for providing 2 valuable input for the writing of the report. R. Einziger (Nuclear Waste Technical Review Board), 3 J. Wang (Oak Ridge National Laboratory), and M. Flanagan

-Bales (U.S. Nuclear Regulatory4 Commission) also provided valuable insights, observations, and recommendations.

5

xiii ABBREVIATIONS AND ACRONYMS 1 ADAMS Agencywide Documents Access and Management System AMP aging management program ANL Argonne National Laboratory ANS American Nuclear Society ANSI American National Standards Institute b width BWR boiling-water reactor C d concentration at d is solution C p concentration at precipitation CFR Code of Federal Regulations cg center of gravity CoC Certificate of Compliance CIRFT cyclic integrated reversible

-bending fatigue tester CRUD C h a lk River unknown deposit CWSR A cold worked stress relieved annealed p/D mo offset strain dp temperature hysteresis (dissolution

-precipitation

) D mi inner (metal) cladding diameter D mo outer (metal) cladding diameter DLF dynamic load factor DTT ductility transition temperature DOE U.S. Department of Energy DSS dry storage system average tensile strain

-N strain per number of cycles E elastic modulus E c elastic modulus of the cladding E p elastic modulus of the fuel pellet EOL end-of-life EPRI Electric Power Research Institute GBC general burnup credit GTCC greater-than-Class-C waste h height h m cladding (metal) thickness HAC hypothethical accident conditions (transportation)

HBR H. B. Robinson HBU high burnup HRT hydride reorientation treatment Hz hertz I moment of inertia I c moment of inertia of the cladding

xiv I p moment of inertia of the fuel pellet IAEA International Atomic Energy Agency IFBA integral fuel burnable absorber ISFSI independent spent fuel storage installation ISG Interim Staff Guidance curvature -N curvature per number of cycles keff k-effective l rod length between spacers LBU low burnup LVDT linear variable differential transformer M bending moment n i number of strain cycles at strain level i N i number of strain cycles to produce failure at i NCT normal conditions of transport NRC U.S. Nuclear Regulatory Commission ORNL Oak Ridge National Laboratory P i rod internal pressure P o rod external pressure PNNL Pacific Northwest National Laboratory PWR pressurized-water reactor r outer radius RCT ring compression testing RHCF radial hydride continuity factor RIP rod internal pressure RXA recrystallized annealed average tensile stress cladding hoop stress z cladding longitudinal stress SNF spent nuclear fuel SRP standard review plan SSC structure, system, and component T d dissolution temperature T p precipitation temperature w uniform applied load ymax distance to the neutral axis

xv Units of Measure C Celsius F Fahrenheit ft foot g 9.806 m/s 2 GWd/MTU gigawatt-days per metric ton of uranium h hour i n. inch lb pound m meter micrometer, 1 x 10-6 meter mm millimeter, 0.001 meter MPa megapascal, 1 x 10 6 pascals N newton N*m newton meter Pa pascal psi pounds per square inch s second Torr Torr (unit of pressure) wppm parts per million by weight

1-1 1 1 1.1 2 As required by Title 10 of the Code of Federal Regulations (10 CFR) 72.44(c), a specific license 3 for dry storage of spent nuclear fuel (SNF) is to include technical specifications that, among 4 other things, define limits on the fuel and allowable geometric arrangements. , Further, a s 5 required by 10 CFR 72.236(a), a Certificate of Compliance (CoC) for a dry storage system 6 (DSS) design must include specifications for the type of spent fuel (i.e., boiling water reactor 7 (BWR), pressurized water reactor (PWR), or both), maximum allowable enrichment of the fuel 8 prior to any irradiation, burn

-up (i.e., megawatt

-days/MTU), minimum acceptable cooling time of 9 the spent fuel before storage in the spent fuel storage cask, maximum heat designed to be 10 dissipated, maximum spent fuel loading limit, condition of the spent fuel (i.e., intact assembly or 11 consolidated fuel rods), and inerting atmosphere requirements. These specifications ensure 12 that the loaded SNF assemblies remain within the bounds of the safety analyses in the 13 approved design basis.

14 The regulations in 10 CFR Part 72, "Licensing Requirements for the Independent Storage of 15 Spent Nuclear Fuel, High

-Level Radioactive Waste, and Reactor

-Related Greater Than Class C 16 Waste," include a number of fuel

-specific and DSS

-specific requirements that may be 17 dependent of the design-basis condition of the fuel cladding. As required by 10 CFR 18 72.122(h)(1), the SNF cladding is to be protected against degradation that leads to gross 19 ruptures or the fuel must be otherwise confined such that degradation of the fuel during storage 20 will not pose operational safety problems with respect to its removal from storage. In addition, 21 10 CFR 72.122(l) states that the DSS must be designed to allow ready retrieval of the SNF. 22 According to Interim Staff Guidance (ISG)

-2, Revision 2, "Fuel Retrievability in Spent Fuel 23 Storage Applications,"

issued in April 2016 (NRC, 2016a).1 This may be demonstratedon an 24 assembly basis per the approved design basis. The condition of the fuel cladding may also 25 impact the safety analyses used to demonstrate compliance with DSS

-specific requirements in 26 10 CFR 72.124(a), 10 CFR 72.128, and 10 CFR 72.236(m).

27 Similarly for transportation, the regulations in 10 CFR Part 71, "Packaging and Transportation of 28 Radioactive Material," also include a number of fuel

-specific and package

-specific requirements.

29 The regulations in 10 CFR 71.31 , "Contents of Application" and 10 CFR 71.33, "Package 30 description," requires an application for a transportation package to describe the proposed 31 package in sufficient detail to identify the package accurately and provide a sufficient basis for 32 evaluation of the package, which includes a description of the chemical and physical form of the 33 allowable contents. The regulations in 10 CFR Part 71 also require that (1) the geometric form 34 of the package contents not be substantially altered under the tests for normal conditions of 35 transport (NCT) (10 CFR 71.55(d)(2))

and (2) a package used for the shipment of fissile material 36 is to be designed and constructed and its contents so limited that under the tests for 37 hypothetical accident conditions (HAC) specified in 10 CFR 71.73, "Hypothetical Accident 38 Conditions," the package remains subcritical (10 CFR 71.55(e). The requirement assumes that 39 1 The current revisions of all ISG documents will be rolled into revised standard review plans (SRPs) for dry storage and transportation of SNF, as appropriate, and will then be removed from the public domain. The revised SRPs will be issued for public comment prior to being finalized.

1-2the fissile material is in the most reactive credible configuration consistent with the 1 damaged condition of the package and the chemical and physical form of the contents 2 (10 CFR 71.55(e)(1)).

3 To comply with the requirements mentioned above, the fuel cladding generally serves a design 4 function in both DSSs and transportation packages for ensuring that the configuration of 5 undamaged and intact fuel remains within the bounds of the reviewed safety analyses.

2 6 Therefore, an application should address potential degradation mechanisms that could result in 7 gross cladding ruptures during operations. To assist the safety review of potential degradation 8 mechanisms, the U.S. Nuclear Regulatory Commission (NRC) staff (the staff) has historically 9 issued guidance on acceptable storage and transport conditions that limit SNF degradation 10 during operations and ensure that the reviewed safety analyses remain valid.

11 1.2 Fuel Cladding Performance and Staff's Review Guidance 12 Time-dependent (i.e., age

-related, not event

-related) mechanisms resulting in changes to the 13 fuel cladding performance are all primarily driven by the fuel's temperature, rod internal 14 pressure (and corresponding pressure

-induced cladding hoop stresses), and the environment 15 during dry storage or transport operations

. Contrary to the hoop stresses experienced by t he 16 fuel cladding during reactor operation, which are generally compressive because of the high 17 reactor coolant pressure , t he hoop stresses during drying

-transfer, dry storage, and transport 18 operations are tensile because of the low pressure external to the cladding. For instance, the 19 pressure of the environm ent surrounding the fuel in the reactor can be 16 MPa (1.2 x 10 5 Torr) 20 while the environment surrounding the fuel in the DSS confinement cavity may be as low as 21 400 Pa (3 Torr) at the end of vacuum drying and 0.5 MPa (3.75 x 10 3 Torr) during dry storage.

22 The magnitude of the cladding hoop stress es will depend on the differential pressure across the 23 cladding wall and thus the rod internal pressure at a given time. Various factors determine the 24 rod internal pressu re , including the fuel's fabrication and irradiation conditions (i.e., fabrication 25 gas fill pressure, cladding thickness, presence of burnable absorbers, burnup) and the average 26 gas temperature within the fuel rods.

The average gas temperature within the fuel rods has a 27 first-order effect on the hoop stress in the cladding and thus cladding performance, and 28 therefore it is critical to controlling the peak cladding temperature of the fuel rods during vacuum 29 drying and storage/transport operations to temperatures demonstrated to preserve cladding 30 integrity. 31 To assist in the safety review of DSS and transportation packages, the staff has developed 32 guidance with a supporting technical bas is for setting adequate fuel conditions, including 33 acceptable peak cladding temperatures during short

-term loading operations so that the 34 cladding meets the pertinent regulations. Historically, guidance has been issued as ISG-11, 35 "Cladding Considerations for the Transportation and Storage of Spent Fuel,"

which has b een 36 revised multiple times to incorporate new data and lessons learned from the staff's review 37 experience. Initial standard review plans (SRPs) prior to ISG

-11 stated that DSSs and 38 transportation packages needed to be dried to a level where galvanic corrosion could be ruled 39 out as a fuel degradation mechanism. The guidance specified moisture levels only for low 40 burnup (LBU) fuel (i.e., burnup below 45 GWd/MTU) because of the lack of degradation data at 41 higher burnup values. In 1999, the staff first issue d ISG-11 to supplement the SRPs by 42 2 If the fuel is classified as damaged, a separate canister (e.g., a can for damaged fuel) that confines the assembly contents to a known volume may be used to provide this assurance.

1-3addressing potential degradation of high burnup (HBU) fuel (i.e., burnup exceeding 1 45 GWd/MTU). 2 A year later, the staff issued ISG-11, Revision 1 to incorporate new data, but also to give the 3 applicant the responsibility for demonstrating that the cladding was adequately protected. ISG-4 11, Revision 1 stated that cladding oxidation should not be credited as load

-bearing in the fuel 5 cladding structural evaluation and also defined a 1

-percent creep strain limit on the cladding.

It 6 also discussed the use of damaged fuel cans for confining fuel with gross ruptures. ISG

-11, 7 Revision 1, accounted only for Zircaloy

-clad fuel rods and not for other advanced cladding alloys 8 (e.g., ZIRLO and M5). 9 In 2002, the staff issued ISG-11, Revision 2, to change the definition of damaged fuel, remove 10 the 1-percent creep strain limit, and discuss criteria to limit hydride reorientation in the cladding.

11 It also made the guidance applicable to all zirconium

-based claddings and all burnup levels.

12 The revision described onerous calculations, dependent on the characteristics of the fuel to be 13 stored, to determine the maximum cladding temperature for the design

-bas is fuel per a justified 14 creep strain limit.

Gruss et al. (2004) discuss in more detail the data used for supporting ISG-15 11, Revision 2. Historically, ISG-11 has not discussed the use of an inert atmosphere to 16 mitigate fuel degradation. Peehs (1998) indicated that air could be used as an atmosphere 17 below 200 °C (392 °F) but later research indicated a lower temperature was necessary.

18 Therefore, ISG-22, "Potential Rod Splitting Due to Exposure to an Oxidizing Atmosphere during 19 Short-Term Cask Loading Operations in LWR or Other Uranium Oxide Based Fuel," issued 20 May 2006 (NRC, 2006), addressed the use of an inert atmosphere for loading operations. 21 In November 2003, the staff issued ISG-11, Revision 3 , "Cladding Considerations for the 22 Transportation and Storage of Spent Fuel

" (NRC, 2003a). The guidance was eventually 23 incorporated into NUREG

-1536, Revision 1, "Standard Review Plan for Spent Fuel Dry Storage 24 Systems at a General License Facility," issued in July 2010 (NRC, 2010), although not yet 25 incorporated into a revision of NUREG

-1567, "Standard Review Plan for Spent Fuel Storage 26 Facilities," issued in March 2000 (NRC, 2000 a) (i.e. the standard review plan for specific 27 licenses under 10 CFR Part 72). ISG-11, Revision 3 replaced the calculation of the maximum 28 cladding temperature per a justified creep strain limit with a generic 400 °C (752 °F) peak 29 cladding temperature limit applicable to normal conditions of storage and transportation, as well 30 as short-term loading operations (e.g., drying, backfilling with inert gas, and transfer of the DSS 31 cask or canister to the storage pad). ISG

-11, Revision 3 also defined a higher short

-term 32 temperature limit applicable to LBU fuel if the applicant demonstrated by calculation that the 33 cladding hoop stress would not exceed 90 MPa (1.3 x 10 4 psi) for the proposed temperature 34 limit. The guidance also defined a generic maximum cladding temperature limit of 570 °C 35 (1,058 °F) for off

-normal and accident conditions applicable to all burnups.

36 I n additi on to creep, I SG-1 1 Revision 3 (NRC, 2003 a), also considered minimizing hy dr ide 37 reorientation. At the time of its issuance, the technical basis discussed in ISG

-11, Revision 3 38 supported the staff's conclusion that hydride reorientation would be minimized by maintaining 39 cladding temperatures below 400 °C (752 °F) and restricting the change in cladding 40 temperatures during drying

-transfer operations to less than 65 °C (117 °F). Th is temperature 41 change limit was based on the temperature drop required to obtain the degree of 42 supersaturation required for the precipitation of radial hydrides in a short thermal cycle 43 (see Section 1.5

.1). Therefore, ISG

-11, Revision 3, states that the cladding should not 44 experience more than 10 thermal cycles, each not exceeding 65 °C (117 °F), which provided 45 assurance that hydride reorientation would be limited.

46 1-4Research results obtained since the ISG

-11, Revision 3, have shown that hydride reorientation 1 c an still occur below the generic 400 °C (752 °F) peak cladding temperature limit (Aomi et al, 2 2008; Billone et al., 2013; Billone et al., 2014; Billone et al., 2015

). To better understand 3 hydride reorientation, both the NRC and the U.S. Department of Energy (DOE) have obtain ed 4 additional data on the performance of HBU SNF cladding with reoriented hydrides to determine 5 if the guidance in ISG

-11, Revision 3, ought to be revised. 6 1.3 Cladding Creep 7 Creep is the time-dependent deformation of a material under stress. The main driving force for 8 cladding creep at a given temperature is the hoop stress caused by internal rod pressure, which 9 results from the fission and decay gases released to the gap between the fuel and cladding (Ito, 10 at al., 2004). Fuel pellet swelling may also result in localized stresses on the cladding due to 11 the mechanical interaction between the cladding and the fuel. Pellet swelling may occur due to

12 (1) the incorporation of soluble and insoluble solid fission products in the fuel matrix, (2) the 13 formation of intra- and intergranular fission gas bubbles, particularly in the hot interior region of 14 a fuel pellet, and (3) the formation of a large number of small gas bubbles in the fine

-grained 15 ceramic structure that builds inward from the outer pellet surface for HBU fuel. If excessive 16 creep of the cladding were to occur during dry storage, it could lead to thinning, hairline cracks, 17 or gross ruptures (Hanson et al, 2012) and potentially compromise the ability to safely retrieve 18 by normal means the HBU fuel on a single

-assembly basis (if required by the design basis). 19 The appendix to ISG-11, Revision 3 (NRC, 2003a) reviewed the data used by the staff to obtai n 20 reasonable assurance that creep will not result in gross ruptures for peak cladding temperatures 21 below 400 °C (752 °F). The fabrication of fuel rods is such that the creep of the cladding is self

-22 limiting. As the cladding creeps, the internal volume of the rod increases and stress decreases.

23 However, as the gas volume within the fuel column increases, the average gas temperature 24 also increases

. The net effect is a slow decrease in pressure and hoop stress with increasing 25 creep strain. The stress also decreases with increasing storage or transport time due to the 26 decrease in internal pressure with decreasing temperature. ISG

-11, Revision 3, concluded th e 27 following:

28 1.deformation caused by creep will proceed slowly over time and will decrease the rod 29 pressure, 30 2.the decreasing cladding temperature also decreases the hoop stress, and this too will 31 slow the creep rate so that during later stages of dry storage, further creep deformation 32 will become exceedingly small, and 33 3.in the unlikely event that a breach of the cladding due to creep occurs, it is believed that 34 this will not result in gross rupture.

35 These conclusions are considered applicable to fuel at all burnups because the relatively small 36 differences in creep rate as a function of materials and burnup are not expected to have a 37 significant impact on the maximum creep strains in the rod. The technical basis in ISG

-11, 38 Revision 3 (NRC, 2003a) has provided reasonable assurance to the staff that creep strains 39 during dry storage and transportation will not result in fuel failures nor compromise the assumed 40 fuel configuration in the safety analyses. However, the staff recognizes the uncertainties 41 associated with extrapoling short-term accelerated test data to extended periods of dry storage.

42 The staff further recognizes the separate

-effects nature of the accelerated creep testing 43 conducted to date, which would not account for potential combined effects with other 44 phenomena occurring during dry storage (e.g., annealing of irradiation hardening, hydride 45 reorientation). Therefore, the staff considers it prudent that long

-term observation of HBU SNF 46 1-5stored in a deployed DSS be used to confirm the conclusions of the accelerated short

-term 1 testing. To aid users in demonstrating adequate creep performance during storage periods 2 beyond 20 years, in June 2016, the staff issued guidance in NUREG-1927, Revision 1 , 3 "Standard Review Plan for Renewal of Specific Licenses and Certificates of Compliance for Dry 4 Storage of Spent Nuclear Fuel" (NRC, 2016b), which discusses the use of an Aging 5 Management Program using a surrogate surveillance and monitoring program to provide this 6 confirmatory long term data.

7 1.4 Effects of Hydrogen o n Cladding Mechanical Performance 8 During irradiation, hydrogen is generated due to water

-coolant corrosion (i.e., oxidation) of the 9 cladding, which diffuses into the zirconium

-based material. As the solubility limit of hydrogen in 10 the cladding is exceeded, circumferential hydrides precipitate (Figure 1-1). The preferential 11 circumferential precipitation of the hydrides during reactor operation results from the texture of 12 cladding, which is determined by the manufacturing process. The number density of these 13 circumferential hydrides varies across the cladding wall due to the temperature drop from the 14 fuel side (hotter) to the coolant side (cooler) of the cladding. When the cladding absorbs 15 significant hydrogen, migration and precipitation of dissolved hydrogen into the coolant side of 16 the cladding can result in the formation of a rather dense hydride rim just below the outer 17 cladding oxide layer. The hydride number density and thickness of this hydride rim depend on 18 cladding design and reactor operating conditions for a given fuel type. For example, fuel rods 19 operated at high linear heat rating to high burnup generally have a very dense hydride rim that 20 is less than 10 percent of the cladding wall thickness.

Conversely, fuel rods operated at low 21 linear heat ratings to high burnup have a more diffuse hydride distribution that could extend as 22 far as 50 percent across the cladding wall.

23 1-6 Figure 1-1 Hydride Content [H] and Distribution (Average, Inner 2/3 Diameter of 1 Cladding) i n HBU SNF Cladding (from Billone et al., 2013) 2 The staff concluded in ISG-11, Rev ision 3 (NRC, 2003a), that the hydride rim, along with any 3 cladding metal oxidized during reactor operation, should not be considered as load bearing 4 when determining the effective cladding thickness for the structural evaluation of the assembly 5 in the DSS or transportation package. However, the staff recognizes that there is no reliable 6 predictive tool available to calculate this rim thickness, which varies along the fuel

-rod length, 7 around the circumference at any particular axial location, from fuel rod to fuel rod within an 8 assembly, and from assembly to assembly. Moreover, recent data generated by Argonne 9 National Laboratory (ANL) have shown that for the full range of gas pressures anticipated during 10 drying and storage, the hydride rim remains intact following slow cooling under conditions of 11 decreasing pressure. The results suggest that hydride rims have some load bearing capacity 12 (Billone et al., 2013; Billone et al., 2014; Billone et al., 2015). These results indicate that it may 13 be appropriate to include the hydride rim in the effective cladding thickness calculation. 14 Therefore, the staff considers acceptable the inclusion of the hydride rim thickness in the 15 calculation of the effective cladding thickness when mechanical test data referenced in the 16 structural evaluation ha ve adequately accounted for its presence. Historically, this has been the 17 case during the review of DSS and transportation packages, as applicants have provided 18 1-7mechanical property data generated from tests with irradiated cladding samples with an intact 1 hydride rim. These data includes test results derived from uniaxial tensile tests or pressurized 2 tube tests of samples that do not have a machined gauge section.

3 Applicants have generally relied on a public database of materials properties for Zircaloy

-4, 4 Zircaloy-2 and ZIRLO to analyze the behavior of as

-irradiated cladding (Geelhood et al, 2008; 5 Geelhood et al, 2013) during dry storage and transportation. Additional data for engineering 6 properties (e.g., yield stress, ultimate tensile stress, and uniform elongation) can be found in the 7 open literature for ZIRLO (Cazalis et al

., 2005; Pan et al

., 2013), Optimized ZIRLO (Pan et al., 8 2013), and M5 (Cazalis et al

., 2005; Fourgeaud et al., 2009; Bouffioux et al

., 2013). The 9 applicant should adequately justify the use of any of these properties for the fuel designs cited 10 for use in the DSS or transportation package application. Any use of mechanical properties 11 from uniaxial

-tension and ring

-expansion tests on cladding specimens with machined gauge 12 sections, where some of the hydride rim would have been inadvertently removed during outer

-13 surface oxide removal, should be adequately justified. The mechanical property data from 14 these specimens are still valuable, but characterization of t he ir remaining rim thickness, post-15 test determination of their hydrogen concentration, or both may be needed. 16 1.5 Hydride Reorientation 17 As discussed in Section 1.4, hydrogen infiltrates the cladding during reactor operation. The 18 excess hydrogen (i.e., hydrogen exceeding the solubility limit in the cladding) precipitates 19 primarily in the circumferential

-axial direction. However, under temperature and stress 20 conditions experienced during vacuum drying and storage/transport operations, some of these 21 hydrides may redissolve and subsequently reprecipitate as new hydrides. During this process, 22 the orientation of these precipitated hydrides may change from the circumferential

-axial to the 23 radial-axial direction. 24 The technical basis discussed in ISG-11, Revision 3 (NRC, 2003a) has supported the staff's 25 conclusion that if peak cladding temperatures are maintained below 400 °C (752 °F) and the 26 pressure-induced hoop stresses in the cladding were maintained below 90 MPa (1.3 x 10 4 psi), 27 then hydride reorientation would be minimized. The database used for this determination (see 28 Figure 3 in Chung, 2004) had a mixture of results from irradiated and non

-irradiated material, 29 high and low hydrogen concentrations, different cladding types, different cooling rates, and 30 other variables. In addition, the methods to determine if there were radial hydrides varied 31 considerably from researcher to researcher. Since the issuance of ISG

-11, Revision 3, 32 research results generated at ANL (Billone et al

., 2013; Billone et al., 2014; Billone et al., 2015) 33 and in Japan (Aomi et al

., 2008) have shown that hydride reorientation can still occur at lower 34 temperatures and stresses than those assumed in ISG

-11, Revision 3. Because of the number 35 of variables involved, the staff agreed that it would not be practical to precisely determine the 36 temperature and stress conditions to prevent reorientation. Rather, t he critical question wa s 37 what effect hydride reorientation would ha ve on the mechanical behavior of the cladding , 38 particularly since the desi gn-bas is structural evaluation of the SNF assembly generally assumes 39 as-irradiated cladding mechanical properties (i.e., properties not accounting for hydride 40 reorientation). If hydride reorientation had an observable effect on the mechanical behavior of 41 the cladding (i.e., it decreased the failure strain limit of the cladding in response to stresses 42 during operations), then the failure limits as defined in the design

-bas is structural evaluations 43 would have to be modified. 44 Because both circumferential and radial hydrides are oriented in the planes parallel to the 45 principal normal tensile stress during bending loading, the staff has expected that HBU SNF 46 1-8fatigue strength and bending stiffness would not be sensitive tothe hydride orientation under1 bendi ng mo ments t hat produc e longitudinal tensil e stresses i n t he rod (T ang et al , 2015)3 2 Experimental confirmation of this expectati on was prudent. Therefore, the NRC and DOE 3 conducted complementary research programs to investigate the cyclic fatigue and bend in g 4 strength performance of HBU SNF cladding in both as-irradiated and reoriented conditions 5 (Wang et al

., 201 6; NRC, 2017). 6 Even with the expectation that hydride orientation would not have a significant impact on the 7 fatigue strength and bending stiffness of HBU SNF under bending moments that produce 8 longitudinal tensile stresses in the rod, the staff expressed concern that hydride orientation 9 could impact the failure stresses and strains under pinch

-type loads. Pinch

-type loads could 10 potentially occur during postulated drop accidents in storage, normal conditions of transport 11 (NCT), or hypothetical accident conditions (HAC) during transportation. The staff was 12 particularly concerned with reduced cladding ductility during the HAC 9-m (30-ft) side drop or a 13 tip-over handling accident, where pinch loads could occur due to rod-to-grid spacer contact, rod-14 to-rod contact, or rod-to-basket contact. If the fuel temperature were to be sufficiently low at the 15 time of the accident, these pinch loads could compromise the analyzed fuel configuration.

16 Thus, research was conducted in the United States and Japan to study the ductility of cladding 17 with reoriented hydrides under diametrically

-opposed pinch loads. Ring compression testing 18 (RCT) was used to assess residual ductility of de-fueled HBU SNF cladding specimens 19 subjected to hydride reorientation (see Section 1.5

.4). Th is testing led to the establishment of a 20 ductility transition temperature (DTT) (i.e., a temperature at which the tested cladding segments 21 were determined to lose ductility relative to as

-irradiated cladding

). Th e following section 22 discuss important parameters affecting the DTT and provides the staff's conclusion on its 23 relevance for future licensing and certification actions involving HBU SNF. 24 1.5.1 Hydride Dissolution and Precipitation 25 During drying

-transfer operations, the cladding temperature increases, which causes some of 26 the circumferential hydrides to dissolve as hydrogen. The amount of hydrogen dissolved 27 depends on the temperature (T d) and increases according to the solubility curve (C d) for 28 zirconium-based alloys (Kammenzind et al

., 1996; Kearns, 1967; McMinn, et al

., 2000).

29 Zirconium-based alloys are materials that can have hydrogen in a supersaturated solution 30 because of the extra energy (strain, thermal) required to precipitate zirconium hydrides in the 31 cladding matrix. This results in a hysteresis in the solubility

-precipitation curves as shown in 32 Figure 1-2. 33 3Hydrides are essentially two

-dimensional features since their thickness is relatively small compared to the other two dimensions. Radial hydrides span in the longitudinal and radial directions, and circumferential hydrides span in the longitudinal and circumferential directions. The bending tensile stresses are in the longitudinal direction.

Therefore, the bending tensile stresses are parallel to the plane of both the radial and circumferential hydrides.

1-9Figure 1-2 Dissolution (C d) and Precipitation (C p) Concentration Curves Based on the 1 Data of Kammenzind et al. (1996) for Non-Irradiated Zircaloy-4 (Zry-4) 2 (Revised Figure 1 from Billone, et al., 2014). Also Shown I s the Best Fit t o 3 the Dissolution Curve (C d) for Zirconium (Zr), Zircaloy-2 (Zry-2), a nd 4 Zircaloy-4, Which Includes the Zircaloy

-2 and Zircaloy

-4 Data Generated by 5 Kearns (1967). t dp = T d - T p Refers to the Temperature Drop Required for 6 Precipitation, where T d and T p are the Corresponding Temperatures Iin the 7 Solubility and Precipitation Curves for the Same Hydrogen Content 8 The solubility curves (C d) plotted in Figure 1-2 indicate that the amount of hydrogen that 9 dissolves increases with increasing temperature, but it is relatively independent of alloy 10 composition and fabricated microstructure (recrystallized annealed (RXA) and cold worked 11 stress relieved annealed (CWSRA)) (Kearns, 1967). Both Kammenzind et al (1996) and Kearns 12 (1967) used diffusion couples, with one sample containing excess hydrogen and the other 13 sample containing essentially no hydrogen, exposed to long annealing times (e.g., 2 days at 14 525 °C (977 °F) and 10 days at 260 °C (500 °F)). As shown in Figure 1

-2 , Kearns' dissolution 15 correlation for Zircaloy

-2 and Zircaloy

-4 is in excellent agreement with the correlation of 16 Kammenzind et al. (e.g., 207 wppm versus 210 wppm at 400 °C (752 °F), and 127 wppm versus 17 133 wppm at 350 °C (662 °F)) and is well within experimental error. In terms of precipitation, 18 dp = T d - T p, where T d and T p are the corresponding temperatures in 19 the solubility and precipitation curves at the same hydrogen content) required for precipitation is 20 approximately 65 °C (117 °F). That is, for irradiated cladding that contains no radial hydrides 21 prior to heating, the 65 °C (117 °F) temperature decrease is necessary to initiate precipitation of 22 0 50 100 150 200 250 300 350 400 450 0 100 200 300 400 500Hydrogen Content (wppm)Temperature (°C)Kammenzind Cd (Zry-4)Kammenzind Cp (Zry-4)Kearns Cd (Zr,Zry-2,Zry-4)DissolutionPrecipitationTdp 1-10radial hydrides.

4 However, if circumferential hydrides are present at the peak cladding 1 temperature, some hydrogen will precipitate by growth of the existing circumferential hydrides 2 during this 65 °C (117 °F) temperature drop because of the lower energy required to grow rather 3 than to initiate precipitation of new hydrides (Colas et al

., 2014). The strain field remaining from 4 the regions of the hydrides that dissolved during heating also facilitates the growth of existing 5 hydrides. 6 McMinn et al. (2000) used a different method (differential scanning calorimetry) to generate an 7 independent data set for dissolution

-precipitation curves per non-irradiated and lightly

-irradiated 8 Zircaloy-2 and Zircaloy

-4 samples with low hydrogen content 60 9 wppm) exposed to temperatures less than 320°C (608 °F). The data show the effects of 10 irradiation (increase in solubility), as well as pre

-annealing time and temperature (decrease in 11 solubility). The increase in hydrogen solubility for irradiated materials is likely the result of 12 hydrogen trapped in irradiation

-induced defects. However, it is not clear yet whether the 13 trapped hydrogen is available for precipitation unless the temperature is high enough to anneal 14 out some of these defects. Extrapolation of the dissolution correlation of McMinn et al. (2000) 15 for non-irradiated cladding alloys gives only 172 wppm of dissolved hydrogen at 400 °C (752 °F) 16 and 102 wppm at 350 °C (662 °F), while the data for irradiated cladding agree quite well with 17 the correlations of Kammenzind et al (1996) and Kearns (1967). The staff considers these two 18 sources to be reasonably representative of dry storage and transportation because the long 19 annealing times used to achieve equilibrium for dissolution are more applicable to drying

-20 storage than the much shorter times used for measurements taken by differential scanning 21 calorimetry. Further, the staff considers these data to provide an upper bound for non

-irradiated 22 cladding and close to a best fit for irradiated cladding.

23 The amount of hydrogen dissolved will depend on the peak cladding temperature during drying

-24 transfer, dry storage, and transport operations. This temperature is typically achieved during 25 vacuum drying, which generally takes about 8 to 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> depending on the DSS or transport 26 package design and loading parameters. Figure 1-2, along with an assessment of the axial 27 hydrogen content of the fuel rods and the peak cladding temperature, can be used to estimate 28 the amount of dissolved hydrogen for a given allowable fuel in a DSS or transportation package. 29 The degree of reorientation will depend on the fuel cladding fabrication process, as well as the 30 cladding hoop stresses and temporal thermal profile of the fuel during operations. The following 31 discussions provide additional details on these parameters.

32 1.5.2 Fuel Cladding Fabrication Process 33 The cladding alloy and corresponding fabrication process are important factors for determining 34 the extent of hydride reorientation. Two predominant cladding microstructures are produced 35 during fabrication of zirconium

-based cladding

CWSRA and RXA. Zircaloy

-4 and ZIRLO are 36 generally CWSRA, whereas Zircaloy

-2 and M5 are RXA. Because hydrides tend to precipitate 37 in the grain boundaries, RXA claddings are more susceptible to hydride reorientation, since 38 these cladding types have a larger fraction of grain boundaries in the radial direction (equiaxed 39 grains) relative to CWSRA claddings (which have more elongated grains).

40 4 This hysteresis is the basis for the guidance in ISG

-11, Revision 3 (NRC, 2003a), to limit repeated thermal cycling (repeated heatup/cooldown cycles) during loading operations to less than 10 cycles, with cladding temperature variations that are less than 65

°C (117 °F) each.

1-11 1.5.3 End-Of-Life Rod Internal Pressures and Cladding Hoop Stresses 1 Most rods are initially backfilled with a pressurized inert helium atmosphere to improve thermal 2 conductivity during irradiation and to decrease the rate of cladding creep

-down onto the fuel.

3 During the course of irradiation, fission gases are generated in the fuel pellets.

Some of the 4 fission gas will be released to the void volume within the fuel column and plenum. The fission 5 gas released is about 1 to 3 percent for PWR fuel rods irradiated under low

-to-moderate 6 conditions up to a burnup of about 45 GWd/MTU, at which point the rate of release increases 7 gradually to about 5 to 7 percent for a burnup of 65 GWd/MTU. For BWR fuel rods, the fission 8 gas release can be in the range of 10 to 15 percent at burnups exceeding 45 GWd/MTU. PWR 9 fuel rods with internal burnable poisons (e.g.

, boron-10 in zirconium

-diboride coating on fuel 10 pellets) can also release decay gases (e.g., helium) within the fuel rod. The pressure of these 11 gases in PWR fuel rods increases with burnup due to the increase in fission gas release, the 12 decay gases released from the burnable poisons, and the decrease in void volume resulting 13 from cladding creep down and fuel swelling.

14 The internal pressure of the rod exerts hoop and axial stresses in the cladding, which increase 15 with burnup because of the increase in internal pressure and the decrease in cladding thickness 16 due to waterside corrosion (i.e., oxidation). For BWR fuels, increased cladding oxidation and 17 hydrogen pickup are observed at burnups exceeding 50 GWd/MTU. In PWRs, hydrogen pickup 18 is usually correlated to the oxide thickness, which varies depending on the alloy. The condition 19 of the fuel as it is removed from the reactor is described more fully in the International Atomic 20 Energy Agency (IAEA) Nuclear Energy Series NF

-T-3.8 (IAEA , 2011). 21 Post-irradiation examination of cladding specimens subjected to representative drying

-transfer 22 and dry storage operations has shown that the degree of radial hydride precipitation is very 23 sensitive to the peak cladding hoop stresses. The range of relevant hoop stresses depends on 24 the range of end

-of-life (EOL) rod internal pressures (RIPs), the range of average gas 25 temperatures within fuel rods during drying and storage, and geometric factors used to convert 26 the pressure differences across the cladding to cladding hoop stresses. The following sections 27 discuss these topics.

28 1.5.3.1 End-Of-Life R od Internal Pressures for Pressurized

-Water Reactor Fuel R ods 29 The publicly

-available database for EOL RIPs for PWR fuel rods is sparse relative to the 30 number of rods that have been irradiated. In addition, the RIP data in this database are for 31 standard fuel rods, mostly those clad in zirconium-tin alloy (Zircaloy-4) with older (1975

-1985) 32 fuel designs and reactor operating conditions. Thus, the database is heavily populated with 33 data from what are generally called "legacy" fuel rods. Figure 1-3 shows the publicly

-available 34 data for standard fuel rods, as collected by the Electric Power Research Institute (EPRI) 35 (Machiels, 2013). Data points are for EOL RIP s extrapolated to 25

°C (77 °F), and are identified 36 by the reactor, the assembly design, and the as

-fabricated helium fill pressure at 25°C (77 °F). 37 Data points labeled as "ENUSA" are for fuel rods irradiated in the Vandellos Unit 2 reactor in 38 Spain. 39 The database consists of 92 data points:

40 -4 and 3 ZIRLO) 41 35 in the range of >45 GWd/MTU to 60 GWd/MTU (25 Zircaloy

-4 and 10 ZIRLO) 42 30 in the range of >60 GWd/MTU to 74 GWd/MTU (15 each of Zircaloy

-4 and ZIRLO

) 43 1-12 Publicly available EOL RIP data are not available for M5-clad SNF rods. Helium fill pressures at 1 fabrication range from 2.00 MPa (290 psi) 3.45 MPa (500 psi). The EOL RIP data appear to 2 be relatively flat between about 40 GWd/MTU and 65 GWd/MTU. 3 Figure 1-3 Publicly-Available Data Collected b y EPRI for PWR End-Of-Life 4 Rod Internal Pressures at 25°C (77 °F) (Reproduction of Figure 2-1 from 5 Machiels (2013))

6 Publicly available data are also not available for ZIRLO-clad integral fuel burnable absorber 7 (IFBA) rods (zirconium diboride

-based), which would have the highest EOL RIP values. Given 8 the sparsity of the database and the absence of publicly available data for standard M5

-clad 9 rods and ZIRLO

-clad IFBA rods, predictions are needed for a wide range of advanced cladding 10 alloys, advanced fuel designs, and more current operating conditions.

11 Two more recent public reports provided EOL RIP values for ZIRLO

-clad IFBA rods from 12 calculations performed with FRAPCON, an NRC

-sponsored fuel performance code.

Oak 13 Ridge National Laboratory (ORNL) published a set of calculations for over 68,000 Zircaloy

-4 14 and ZIRLO fuel rods irradiated during the first 10 cycles of the Watts Bar Nuclear Plant Unit 1 15 reactor (Bratton et al, 2015). FRAPCON was used to predict RIPs for standard rods and IFBA 16 rods irradiated for one cycle, two cycles, and three cycles, with each cycle consisting of 17 18 months. ORNL modified the FRAPCON code to account for the helium release from the 18 IFBA rods. Figure 1-4 shows the results extrapolated to 400 °C (752 °F), which show the end-19 of-cycle RIP values for the IFBA rods were calculated to be higher than the values for 20 standard rods.

21 1-13 Figure 1-4 Calculated Rod Internal Pressures for the First 10 Cycles o f the Watts Bar 1 Nuclear Plant Unit 1 Reactor Under Vacuum Drying Conditions (P o = 133 Pa 2 (1 Torr); Tcladding(Max) = 400°C (752 °F) (Reproduction of Figure 4.24 from 3 Bratton et al. (2015)). The Data Points Shaded in Blue are for Standard Fuel 4 Rods (Approximately 21,000 Rods). The Data Points Shaded in Red are for 5 IFBA Fuel Rods (Approximately 47,000 Rods), with t he Lower Values for 6 IFBA Fuel with Annular Blanket Pellets and the Higher Values for Solid 7 Blanket Pellets 8 ORNL's calculated end-of-cycle RIPs from Figure 1

-4 were extrapolated to 25 °C (i.e. the 9 temperature for EPRI's EOL RIP data) to allow comparison with the results in Figure 1

-4. 10 Figure 1-5 shows t he aggregated RIP data at 25 °C (77 °F) within the burnup range of 40 to 11 74 GWd/MTU, along with best-fit and 3- Within the relevant burnup range of 40 12 to 62 GWd/MTU, the average +/- 3

-ues are 4.0

+/- 1.0 MPa ((5.8 +/- 1.4) x 10 2 psi) for standard 13 fuel rods with Zircaloy

-4 and ZIRLO cladding alloys. The 3

-(7.3 x 1 0 2 psi) is 14 a reasonable upper bound of data and calculated values for standard fuel rods. Furthermore , 15 reasonable upper bounds for data on IFBA rods are 7.9 MPa (1.1 x 10 3 psi) at 25 °C (77 °F) for 16 fuel designs with hollow (annular) pellets and 10.0 MPa (1.5 x 10 3 psi) for fuel designs with solid 17 pellets. 18 1-14 Figure 1-5 Aggregated Measured and Calculated Values for End-Of-Life Rod Internal 1 Pressures for PWR Fuel Rods.

Pressures are Presented at 25 °C (77 °F) 2 More recently, Richmond and Geelhood (2017) used FRAPCON to calculate EOL RIP for three 3 modern fuel designs with three representative dry storage thermal transients, each involving 4 drying operations with a peak cladding temperature of 400

°C (752 °F). The analyses 5 characterized the effects of fuel design to determine reasonably bounding cladding hoop 6 stresses.

The report provide s results for maximum EOL RIP for IFBA rods to be limited to 7 10.6 MPa (1.5 x 10 3 psi). These results suggest that, even at a peak cladding temperature of 8 400 °C (752 °F), the maximum cladding hoop stresses remain below 90 MPa (1.3 x 10 4 psi) for 9 the bounding ZIRLO-clad IFBA rods. At these pressures, the extent of hydride reorientation will 10 be very limited, if observed at all, which would indicate that the mechanical properties from the 11 as-irradiated condition will remain unchanged. The staff recognizes the discrepancies in the 12 results between Bratton et al

. (2015) and Richmond and Geelhood (2017) and is evaluating the 13 merits of both approaches used for the FRAPCON analyses. Until that evaluation is complete, 14 since the data of Bratton et al (2015) results in the highest EOL RIP and corresponding hoop 15 stresses, the staff will assume those values to be conservative and bounding when evaluating 16 the potential of hydride reorientation.

17 1.5.3.2 Gas Temperatures for Fuel Rods During Drying-Transfer , Storage And 18 Transportation 19 Peak temperatures for the gas inside fuel rods are highly dependent on the DSS or 20 transportation package design, fuel system design, fuel burnup, operating history, package 21 loading density, and the length of time the fuel was cooling in the spent fuel pool. Figure 1-6 22 0 1 2 3 4 5 6 7 8 9 10 11 40 45 50 55 60 65 70 75EOL PWR RIP at 25

°C (MPa)Rod Average Burnup (GWd/MTU)Data: Standard RodsFRAPCON: Standard RodsFRAPCON-IFBA-HBPFRAPCON-IFBA-SBPData Average + 3

-Data Average 1-15 shows the extrapolation of the upper-bound pressures shown in Figure 1-5 to rod internal gas 1 temperatures that may be experienced during drying

-transfer, dry storage and transportation.

2 The data are presented for the relevant burnup range of 45 to 62 GWd/MTU.

3 Figure 1-6 Calculated Rod Internal Pressure (Average +

3; 45 To 62 Gwd/MTU) a s a 4 Function o f the Spatially Averaged Gas Temperature for PWR Fuel Rods 5 (i.e., Standard Rods

), ZIRLO-Clad IFBA Rods with Hollow (Annular) Blanket 6 Pellets, a nd ZIRLO-Clad IFBA Rods with Solid Blanket Pellets 7 1.5.3.3 Peak Cladding H oop S tresses for Pressurized

-Water Reactor Fuel Rods During 8 Drying-Transfer a nd Storage/Transport Operations 9 ) is a function of the gas pressure difference across the cladding 10 wall (P i - P 0), where P i is the internal rod pressure and P o is the external pressure

, the cladding 11 inner diameter (D mi), and the cladding metal wall thickness (h m), as shown in Eqn.

1-1 for the 12 average hoop stress across the cladding wall (Figure 1-7). Gap closure is not considered.

13 = [D mi / (2

• h m)] (P i - P o) (Eqn. 1-1) 14 The geometrical parameter D mi/(2

• h m) will tend to increase with burnup due to waterside 15 corrosion of the cladding outer surface, which reduces h

m. Nominal as

-fabricated values of D mi 16 and h m for standard 17 x 17 PWR fuel rods are 8.36 mm and 0.57 mm, respectively, which 17 gives a geometrical factor of 7.33. However, fuel rod cladding is manufactured to specifications 18 with tolerances (i.e., +/- values) for the outer diameter and inner diameter and a minimum wall 19 thickness (e.g., >

0.56 mm). Thus, even for fresh fuel rods, the geometrical parameter will vary 20 somewhat along the length of one fuel rod and from fuel rod to fuel rod. For LBU SNF, the 21 decrease in h m is small and is partially counteracted by the decrease in D mi due to creep down.

22 4 6 8 10 12 14 16 18 20 22 25 50 75 100 125 150 175 200 225 250 275 300 325 350Pressure (MPa)Average Rod Internal Gas Temperature

(°C)Standard RodsIFBA (Hollow Blanket Pellets)IFBA (Solid Blanket Pellets) 1-16 For HBU SNF with creep out resulting from fuel-cladding mechanical interaction, D mi 1 approaches its as

-fabricated value and h m continues to decrease with burnup.

2 Figure 1-7 Fuel Cladding Tube with Stress Element Displaying Hoop Stress (), 3 Longitudinal Stress (z), Internal Pressure (P i), Cladding Thickness (H m), 4 External Pressure (P o), Circumferential Coordinate (, and Inner Cladding 5 Diameter (D mi) 6 Example calculations may be performed assuming that the EOL inner cladding diameter (D mi) is 7 8.36 mm (i.e., the as-fabricated value) with oxide layer thicknesses of 10 µm (e.g., M5), 40 µm 8 (e.g., ZIRLO), and 80 µm (e.g., Zircaloy-2, Zircaloy-4). The corresponding values of the 9 geometrical factor for these corrosion layers are: 7.41, 7.64, and 7.79.

These factors are used 10 in Eq n. 1-1 to generate the cladding hoop stress versus average gas temperature for several 11 values of corrosion layer thickness. The value of P o (varies from approximately 4 x 10-4 MPa 12 (3 Torr) for vacuum drying to less than 0.5 MPa (3.8 x 10 3 Torr) for storage) is assumed to be 13 zero for these calculations. Figure 1-8 shows that the hoop stress is a strong function of the 14 average gas temperature and a weaker function of the corrosion

-layer thickness. Calculated 15 cladding hoop stress values varied from 57 MPa (8.3 x 10 3 psi) at 180 °C (356 °F) for the 10

-µm 16 oxide case to 83 MPa (1.2 x 10 4 psi) at 340 °C (644 °F) for the 80

-µm oxide case.

17 1-17 Figure 1-8 Calculated Values of Cladding Hoop Stress (Average + 3; 45 To 62 1 Gwd/MTU) Vs. the Spatially

-Averaged Internal Gas Temperature for 2 Standard 17x17 PWR Fuel Rods With Corrosion Layers of 10 µm, 40 µm, 3 and 80 µm 4 For ZIRLO-clad IFBA rods, some 17 x 17 designs use smaller diameter cladding 5 (9.14 mm (0.360 in.)) and comparable cladding wall thickness (0.572 mm (0.023 in.)). This 6 design is called the "optimized fuel assembly." The geometrical factor for converting pressure to 7 hoop stress is 7.00 for as

-fabricated cladding and 7.28 for cladding with a 40

-µm corrosion 8 layer, which assumes the cladding inner diameter (8.00 mm (0.315 in.) remains the same as the 9 as-fabricated inner diameter for HBU SNF. Figure 1-9 shows that the cladding hoop stress for 10 the 40-µm corrosion

-layer case increases with gas temperature from about 90 MPa 11 (1.3 x 10 4 psi) to 120 MPa (1.7 x 10 4 psi) for IFBA fuel with hollow blanket pellets and from 12 about 110 MPa (1.6 x 10 4 psi) to 150 MPa (2.2 x 10 4 psi) for IFBA fuel with solid blanket pellets.

13 35 40 45 50 55 60 65 70 75 80 85 25 50 75 100 125 150 175 200 225 250 275 300 325 350PWR Rod Hoop Stress (MPa)Average Rod Internal Gas Temperature (°C) m oxidem oxidem oxide 1-18 Figure 1-9 PWR Hoop Stress a s a Function of Internal Gas Temperature for 17 x 17 1 IFBA Fuel Rods (for Both Hollow Blanket and Solid Blanket Pellets) with a 2 40-µm Corrosion Layer 3 The above discussion provide s a technical basis for the staff to defin e conservative bounding 4 cladding hoop stress conditions for the testing of HBU SNF mechanical performance, as the 5 assessment in Chapter 2 will discuss

. 6 1.5.4 Ring Compression Testing 7 Ring compession testi ng (RCT) has been conducted in the United States and Japan to assess 8 residual ductility of cladding with reoriented hydrides following pinch loads (Aomi et al

., 2008; 9 Billone et al., 2013; Billone et al., 2014; Billone et al., 2015). RCT of zirconium

-based cladding 10 alloys has shown reduced ductility when subjected to pinch loads at a sufficiently low 11 temperature; this temperature has been generally referred to as a ductile

-to-brittle transition 12 temperature or ductility transition temperature (DTT)

. 13 In previous NRC-sponsored research, Argonne National Laboratory (ANL) sectioned rings from 14 pressurized

-and-sealed rodlets fabricated with cladding from ZIRLO

-clad and Zircaloy clad 15 fuel rods irradiated to high burnup (beyond the NRC's peak rod licensing limit i n 16 commercial PWRs) (Billone et al., 2013)

(Figure 1-1 0). These rodlets had been heated to a 17 peak temperature of 400 °C (752 °F) (consistent with the guidance limit in ISG

-11, Revision 3 18 (NRC, 2003 a) and held at this temperature for 1 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> with variable target hoop stresses 19 (110 MPa (1.6 x 10 4 psi), 140 MPa (2.0 x 10 4 psi)), and then slow-cooled at 5

°C/h (9 °F/h) 20 under conditions of decreasing pressure and hoop stress. Metallographic examination of one 21 cladding ring surface per rodlet was used to quantify the degree of radial hydride precipitation in 22 terms of the average length of radial hydrides. Several other rings were used to determine the 23 40 60 80 100 120 140 160 25 50 75 100 125 150 175 200 225 250 275 300 325 350PWR Rod Hoop Stress (MPa)Average Rod Internal Gas Temperature (°C) IFBA (Hollow Blanket Pellets)IFBA (Solid Blanket Pellets) 1-19 average hydrogen content of the rodlet, along with circumferential and axial variations in 1 hydrogen conten t. Up to four rings were subjected to RCT to induce pinch loads at test 2 temperatures from 20

°C (68 °F) to 200 °C (392 °F). 3 Figure 1-10 RCT of a Sectioned Cladding Ring Specimen in ANL's Instron's 8511 Test 4 Setup. Tests Were Conducted in the Displacement

-Controlled Mode to a 5 1.7-Mm Maximum Displacement in a Controlled Temperature Environment (6 p = RCT Offset Displacement at 12 O'clock Position Relative to Static 7 Support at 6 O'clock

D mo = Outer Diameter of Cladding Metal

p/D mo = 8 RCT Offset Strain (Percent)) (Reproduction Of Figure 6 From Billone et al., 9 2012)) 10 RCT load-displacement curves were used to determine the offset displacement (normalized to 11 the pretest sample outer diameter to give offset strain) as a function of test temperature. The 12 offset strain was plotted against test temperature for each rodlet to determine the DTT 13 (see Figure 1-1 1). Post-RCT metallographic examinations were also performed to determine 14 the number and extent of cracks that had formed, as well as to generate additional data for the 15 degree of radial hydride precipitation (Billone, et al., 2013).

16 To define ductility per RCT load-displacement data, a 2

-percent offset strain (p/D mo) before a 17 crack extend ed through more than 50 percent of the cladding wall thickness was chosen as the 18 figure of merit for the transition between ductile and brittle behavior (Billone et al., 2013). Figure 19 1-1 1 shows representative deformation (i.e., offset strain) curves as a function of the alloy, peak 20 hoop stress at a 400 °C (752 °F) peak cladding temperature, and actual RCT temperature. The 21 figure also shows the radial hydride continuity factor (RHCF), which represents the effective 22 radial length of continuous radial

-circumferential hydrides normalized to the wall thickness. ANL 23 used the RHCF as a figure of merit f o r determining the degree and severity of radial hydrid e 24 precipitation. The radial hydrides in Zircaloy

-4 HBU SNF ring specimens were relatively short 25 (i.e., RHCF of 9 percent for a peak hoop stress of 110 MPa (1.6 x 10 4 psi), and 16 percent for a 26 peak hoop stress of 140 MPa (2.0 x 10 4 psi)) and the ductility increased gradually with 27 temperature. In ZIRLO-clad HBU SNF ring specimens

, the radial hydrides were longer (i.e., 28 RHCF of 30 percent for a peak hoop stress of 110 MPa (1.6 x 10 4 psi), and 65 percent for a 29 1-20 peak hoop stress of 140 MPa (2.0 x 10 4 psi)) and the ductility increased sharply with the 1 increase in RCT temperature. ANL fit th e limited ZIRLO data points with S

-shaped curves 2 (hyperbolic tangent functions) typical of materials that exhibit a ductile

-to-brittle transition. The 3 data show that the DTT shifted from around room temperature in a cladding material with short 4 radial hydrides to higher values in a cladding material with longer radial hydrides. The limited 5 data also indicates a trend of lower DTTs for materials with lower peak cladding stresses. 6 Figure 1-11 Ductility vs. RCT for Two PWR Cladding Alloys Following Slow Cooling 7 from 400°C (752 °F) at Peak Target Hoop Stresses of 110 Mpa (1.6 x 10 4 Psi) 8 and 140 Mpa (2.0 x 10 4 Psi) (From Billone e al., 2013) 9 ANL also conducted RCT research under DOE sponsorship.

It obtained results for the following 10 (Billone et al., 2014; Billone et al., 2015)

11 HBU Zircaloy-4 in the as

-irradiated condition with moderate

-to-high hydrogen content 12 HBU ZIRLO in the as

-irradiated condition and following simulated drying

-storage at peak 13 temperatures of 400

°C (752 °F) and 350 °C (662 °F) with peak hoop stresses from 14 80 MPa (1.2 x 10 4 psi) to 94 MPa (1.4 x 10 4 psi) 15 HBU M5 in the as

-irradiated condition and following simulated drying

-storage at 16 400 °C (752 °F) with peak hoop stresses of 90 M Pa (1.3 x 10 4 psi), 110 MPa 17 (1.6 x 10 4 psi), and 140 MPa (2.0 x 10 4 psi) 18 ANL conducted two additional tests with HBU ZIRLO cladding subjected to three drying cycles 19 (e.g., from 400 °C (752 °F) to 300 °C (572 °F) and from 350

°C (662 °F) to 250 °C (482 °F)) at 20 peak hoop stress of about 90 MPa (1.3 x 10 4 psi). The latter results suggest that multiple drying 21 0 2 4 6 8 10 12 14 0 25 50 75 100 125 150 175 200 225Offset Strain (percent)RCT Temperature (°C)9+/-5% RHCF; Zry-4 @ 110 MPa16+/-4% RHCF; Zry-4 @140 MPa 30+/-12% RHCF; ZIRLOŽ @ 110 MPa 65+/-17% RHCF; ZIRLOŽ @140 MPa Brittle Ductile 1-21 cycles ha ve no effect on the length of radial hydrides or the DTT at this low stress level.

Figur e s 1 1-1 2 through 1-1 4 show results for Zircaloy

-4 , ZIRLO , and M5 in both as

-irradiated and hydride

-2 reoriented condition following cooling from 400°C (752 °F) (Billone et al., 2014; Billone et al., 3 2015). 4 Figure 1-12 Ductility Data, as Measured by RCT , for A s-Irradiated Zircaloy-4 and 5 Zircaloy-4 Following Cooling from 400 °C (752 °F) Under Decreasing 6 Internal Pressure and Hoop Stress Conditions (From Billone e t al., 2013) 7 0 2 4 6 8 10 12 14 0 25 50 75 100 125 150 175 200 225Offset Strain (percent)RCT Temperature (°C)300+/-15 wppm H640+/-140 wppm H520+/-90 wppm H615+/-82 wppm HBrittleDuctile145 MPaat 400°C113 MPaat 400°CHigh Burnup Zircaloy-4 As-Irradiated 1-22 Figure 1-13 Ductility Data, as Measured by RCT , for a s-Irradiated ZIRLO a nd ZIRLO 1 Following Cooling from 400 °C (752 °F) Under Decreasing Internal Pressure 2 and Hoop Stress Conditions (From Billone e t al., 2013) 3 0 2 4 6 8 10 12 14 0 20 40 60 80 100 120 140 160 180Offset Strain (percent)RCT Temperature (°C)530+/-70 wppm H535+/-50 wppm H530+/-115 wppm H480+/-131 wppm H385+/-80 wppm HBrittleDuctile111 MPaat 400°C80 MPaat 400°CHigh Burnup ZIRLO As-Irradiated89 MPaat 400°C 1-23 Figure 1-14 Ductility Data , as Measured by RCT , for As-Irradiated M5 a nd M5 Following 1 Cooling from 400 °C (752 °F) Under Decreasing Internal Pressure and Hoop 2 Stress Conditions (From Billone et al., 2013

) 3 The staff recognizes the uncertainties associated with the ductility curve fits of ANL's RCT data 4 because of the limited number of data points. However, the limited results appear ed to support 5 the following general conclusions: (1) the DTT generally increases with increasing hoop 6 stresses (i.e., the ductility transition shifts to higher cladding temperature), (2) both the 7 susceptibility to radial hydride precipitation and ductility changes depend on cladding type and 8 initial hydrogen content, and (3) depending on the cladding and test conditions, the DTT can 9 occur at temperatures in the range of 20 °C (68 °F) to 185 °C (365 °F). The results for as

-10 irradiated Zircaloy

-4 are consistent with studies by Wisner and Adamson (1998) and Bai et al 11 (1994). The staff considered these conclusions when defining limiting conditions for inducin g 12 radial hydrides and conducting fatigue and bending testing of HBU SNF (see Chapter 2).

13 It is important to note that th e DTT is not an intrinsic property of a cladding alloy material with a 14 given homogeneous composition, in the classical metallurgical sense, but it is highly dependent 15 on the composite microstructure (hydride

-zirconium matrix, as determined by reactor operating 16 conditions

), fabrication conditions (degree of cold working, recrystallization) and the operating 17 conditions during drying

-transfer, storage or transportation (peak cladding temperature, peak 18 hoop stress, temporal cooling profile

). Further, the DTT was established based on an arbitrarily

-19 defined performance criterion (e.g., 50 percent cladding through

-wall crack prior to 2

-percent 20 offset strain deformation), and based on a limited number of data points for each cladding alloy.

21 It is also important to note that, due to the radial and axial temperature gradients in a DSS or 22 transportation package, it is highly likely that only a small fraction of the cladding in a given 23 assembly will reach high enough temperatures and hoop stress es to have sufficient hydride 24 reorientation during cooling. Those hotter axial locations of the cladding will likely be the last to 25 reach a DTT during transpor

t. 26 0 2 4 6 8 10 12 14 0 25 50 75 100 125 150 175 200 225Offset Strain (%)RCT Temperature (°C)76+/-5 wppm H58+/-15 wppm H (90 MPa)72+/-10 wppm H (111 MPa)94+/-4 wppm H (142 MPa)142 MPa at 400°C111 MPaat 400°CDuctileBrittleHigh Burnup M5 As-Irradiated90 MPaat 400°C 1-24 1.5.5 Staff's Assessment of Ring Compression Testing Results 1 As previously discussed, the staff has long expected that hydride reorientation would not 2 compromise cladding integrity due to fuel rod bending (i.e., bending expected during normal 3 conditions of storage and transport

), since the principal tensile stress field associated with rod 4 bending caused by lateral inertia loads is parallel to both radial and circumferential hydrides 5 (Tang et al

., 2015). The staff has considered that any reduced cladding ductility due to hydride 6 reorientation could only potentially compromise the analyzed fuel configuration for pinch loads 7 experienced during drop accident scenarios, if the fuel ha d significantly cooled during the 8 transportation period. More specifically, the staff had expressed concern that reorientation 9 could decrease failure stresses and strains in response to transportation

-induced pinch loads 10 during a 9-m (30-ft) drop scenario as a result of rod-to-grid spacer contact, rod-to-rod contact, or 11 rod-to-basket contact. 12 To address the concern of reduced ductility during drop accidents, the staff previously proposed 13 varied approaches to demonstrate that the failure limits for as

-irradiated cladding as used in the 14 design-bas is structural evaluations would continue to be adequate even if hydride reorientation 15 occurred. One of these approaches was based on justifying an RCT-measured DTT for each 16 cladding alloy in the proposed fuel contents, and demonstrating that the minimum cladding 17 temperature remained above the RC T-measured DTT for the entire duration of transport. The 18 minimum cladding temperature assumed for transport operations would need to be bounding to 19 the contents upon consideration of the cold temperature requirement in 10 CFR 71.71(c)(2), i.e.

20 an ambient temperature of

-40 °C (-40 °F) in still air and shade. If these conditions were met, 21 then mechanical properties of the as

-irradiated cladding material (i.e., material that did not 22 account for the precipitation of radial hydrides

), would be considered adequate for the structural 23 evaluation

. 24 As an alternative approach, if the applica nt could not reasonably demonstrate that sections of 25 the fuel cladding remained above the RCT-measured DTT during the entire duration of 26 transport, the staff propose d that the application provide additional safety analyses assuming 27 hypothetical reconfiguration of the HBU fuel contents. If neither of these two approaches i s 28 satisfactory for demonstrating compliance with 10 CFR Part 71 regulations,then the staff would 29 expect that t he fuel would be canned an d classified as damaged

. 30 Since proposing these approaches, the staff has reevaluated whether results from RCT of 31 defueled specimens are accurately representative or if they are overly conservative relative to 32 the actual hoop

-loading conditions experienced by the fuel during a 9-m (30-ft) drop. During 33 RCT, the circumferential (hoop) tensile bending stress is perpendicular to the plane of the radial 34 hydrides, which is different from the relative orientation of the applied stress and hydrides under 35 axial tensile bending where the longitudinal (axial) tensile bending stress is always parallel to 36 the plane of both the circumferential and radial hydrides. The orientation of the tensile stress is 37 expected to make a difference in the response of the cladding.

38 The RCT defined a D TT used to determine cladding failure due to pinch loads. However, it is 39 necessary to consider the importance of this failure mode in the determination of cladding 40 integrity in the event of a drop accident. To do this, the RCT must be examined for what it is, a 41 test in which diametrically

-opposed concentrated compressive forces are applied to a fuel 42 cladding longitudinal segment that does not contain fuel. During NCT and HAC side drops, the 43 fuel rod is loaded by lateral inertia loads that are resisted by distributed loads applied to the 44 bottom of the rod at the flexible grid spacer springs (Figure 1-1 5). Further, the inertia load in the 45 1-25 rod is transferred to the grid spacer support as a shear force in the cladding (and pellets) not as 1 a concentrated load at the top of the rod.

2 Single Rod Model Single Assembly Model Figure 1-15 Geometric Models for Spent Fuel Assemblies i n Transportation Packages 3 (Reproduction , i n Part, Of Figure 10 from Sanders e t al., 1992) 4 Given that the forces and displacements in the RCT are measurably different from the actual 5 forces and displacements applied to the rod at the grid spacer support, it is not likely that the 6 pinch-mode of failure will play a significant role in undermining cladding integrity. To quantify 7 the difference between these loading cases, the staff analyzed two ring segments for different 8 loading conditions and the change in diameter calculated. In the first case the ring segment 9 was loaded by diametrically

-opposed compressive forces like those of RCT (Case 1, Table 17, 10 Roark and Young (1975)). In the second case the ring segment was supported at the bottom by 11 a concentrated reaction and loaded by a downward load uniformly distributed around the 12 circumference of the ring to simulate a shear loading as in a side drop (Case 13, Table 17, 13 Roark and Young (1975)). In both cases the total applied load was the same.The ratio of the 14 change in diameter of the second case to the first case is 0.48. Thus, the diametrically

-opposed 15 compressive forces produced more than twice the displacement when compared to the 16 circumferentially distributed load. In addition, at the pellet

-cladding interface

, the pellet and 17 cladding are bonded and, thus a gap cannot exist between the

m. Thus , the staff considers that, 18 under a pinch load, ovalization of the cladding cross

-section is very unlikely and any 19 circumferential bending stress that does exist will be negligible. The RCT conducted to dat e 20 does not account for the rod's resistance to ovalization provided by the pellet.

21 Based on the RCT load-displacement data, ANL defined a figure of merit for cladding ductility 22 (i.e., the transition between ductile and brittle behavior) to be a 2-percent offset strain prior to 23 a crack extending through more than 50 percent of the cladding wall (Billone et al., 2013). If 24 the strains experienced during RCT's diametrically

-opposed loads result in twice those that 25 would be experienced during lateral inertial loads, then the DTT is likely to shift to lower 26 temperatures (potentially room temperature or lower). Therefore, the staff considers that the 27 DTT defined by RCT experiments is overly conservative and not representative of actual fuel 28 and stress conditions during NCT and HAC drop scenarios. The DOE is planning on 29 sponsoring a research program in which 25 HBU fuel rods will undergo testing to determine 30 their characteristics, material properties, and rod performance following representative drying

-31 transfer and cooldown (Hanson et al

., 2016). The staff expects that material property testing 32 conducted under this program will provide confirmation that the cladding displacements 33 experienced by fueled cladding specimens during RCT will be lower than those measured in 34 1-26 defueled specimens and that ductility during accident drop scenarios is not compromised.

1 Results from the static and fatigue bend testing discussed in Chapter 2 further justify the 2 staff's conclusion that the pellet imparts structural support to the mechanical performance of 3 the fuel rod.

4 2-1 2 ASSESSMENT OF STATIC BENDING AND FATIGUE STRENGTH 1 RESULTS ON HIGH BURNUP SPENT NUCLEAR FUEL 2 2.1 Introduction 3 The sealed canister or cask cavity serves as the primary barrier in a dry storage system (DSS) 4 or transportation package for protecting against the release of radioactive solid particles or 5 gases from the loaded spent nuclear fuel (SNF) to the atmosphere. The spent fuel cladding 6 also serves as a confinement or containment barrier for preventing radioactive solid particles 7 and fission gasses from being released into the interior cavity of the DSS or transportation 8 package. The cladding not only provides a barrier for preventing the release of radioactive 9 material but also prevents fuel reconfiguration during storage and transport operations.

10 Therefore, the integrity of the cladding is an essential component of a defense

-in-depth strategy 11 to protect the public health and safety.

12 Until recently, research to understand the structural behavior of spent fuel rods during 13 transportation and storage has focused entirely on obtaining mechanical and strength properties 14 of spent fuel cladding. As a result, the flexural rigidity and structural response of fuel rods 15 during normal and accident events ha ve been based on the mechanical and strength properties 16 of only the cladding. The contribution of the fuel pellets to increasing the flexural rigidity of the 17 rod has been neglected. However, recent research discussed in NUREG/CR

-7198 , Revision 1 , 18 "Mechanical Fatigue Testing of Hig h-Burnup Fuel for Transportation Application," issued 19 October 2017 (NRC, 2017a), on the static bending response and fatigue strength of fuel rods 20 considered as a composite system of cladding and fuel pellets, has begun to provide some of 21 the necessary data to allow a more accurate assessment of the structural behavior of the 22 composite fuel rod system under normal conditions of transport (NCT) and hypothetical accident 23 conditions (HAC), as well as DSS drop and tip

-over events.

24 2.2 Cyclic Integrated Reversible Fatigue Tester 25 In 2009, the U.S. Nuclear Regulatory Commission (NRC) tasked Oak Ridge National Laboratory 26 (ORNL) with investigating the flexural rigidity and fatigue life of high burnup (HBU) SNF 27 (N RC, 2017a). The testing was designed to evaluate the fuel rod as a composite system, 28 including the presence of intact fuel inside the cladding and any pellet/cladding bonding effects.

29 The project proceeded in two phases. Phase I involved testing HBU SNF in the as-irradiated 30 state, where hydrides are expected to be predominantly in the circumferential

-axial orientation.

31 Phase II involved testing HBU SNF segments subjected to a treatment designed to reorient the 32 hydrides in the cladding to be predominantly in the radial

-axial orientation. All testing was 33 conducted at room temperature, which is expected to result in the most limiting cladding 34 ductility.

35 In response to the NRC tasking, in 2011, ORNL proposed a bending fatigue system for testing 36 HBU SNF rods. The system is composed of a U

-frame equipped with load cells for imposing 37 pure bending loads on the SNF rod test specimen and measuring the in-situ curvature of the 38 fuel rod during bending using a set

-up of three linear variable differential transformers (LVDT) 39 (Figure 2-1). Pure bending is a condition of stress in which a bending moment is applied to a 40 beam without the simultaneous presence of axial, shear, or torsional forces.

41 2-2 Figure 2-1 Horizontal Layout of ORNL U-Frame Setup (Top), Rod Specimen and Three 1 Lvdts for Curvature Measurement (Mid dle), and Front View of CIRFT 2 Installed in ORNL Hot Cell (Bottom) (Figure 4 from NUREG/CR-7198, 3 Revision 1 (NRC, 2017a

)) 4 SNF rod End-blocks LVDT clamp Three LVDTs for curvature measurement Rod specimen Rigid arms Connecting plates Universal testing machine links Load cell (middle) (bottom) (top) 2-3 On August 19, 2013, a testing system was installed in a hot cell at ORNL's Irradiated Fuels 1 Examination Laboratory and formally named the "cyclic integrated reversible

-bending fatigue 2 tester" (CIRFT). After tuning of the test system and performance of benchmark testing in 3 September 2013, Phase I testing began on HBU SNF rod segments with intact Zircaloy-4 4 cladding irradiated in the H.B. Robinson Steam Electric Plant (HBR). ORNL completed four 5 static tests under displacement control at the rate of 0.1 mm/s to a maximum displacement of 6 12.0 mm. In early November 2013, the benchmark and static test results were critically 7 reviewed at a meeting between representatives from the NRC and ORNL. Dynamic testing was 8 then initiated, and 16 cyclic tests were completed in the Irradiated Fuels Examination 9 Laboratory

. Load ranges applied to the CIRFT varied , to produce bending moments in the rod, 10 from +/-5.08 to +/-35.56 N*m. There were 12 dynamic tests with rod fracture and 4 tests without 11 rod fracture

. One of the cyclic tests reached 1.3 x 10 7 cycles with no rod fracture. The test was 12 terminated as higher cycles would not be expected during actual transport.

13 Phase II testing began in 2016, again using HBR HBU SNF rods with intact Zircaloy

-4 cladding, 14 which had been subjected to an aggressive hydride reorientation treatment (HRT)

(see 15 Section 2.3.4). ORNL completed testing on four specimens in the CIRFT following an HRT: 16 one in static loading (hereafter referred to as HR2)

, and three in dynamic loading (hereafter 17 referred to as HR1, HR3 , and HR4). The fatigue lifetime and flexural rigidity of these samples 18 were compared to the results obtained in Phase I for as-irradiated samples.

19 The following observations can be made about the results of the static testing:

20 The HBR HBU SNF rods in the as

-irradiated state exhibited a multiple

-stage constitutive 21 response, with the two linear stages followed by a nonlinear stage. The flexural rigidity at 22 2, corresponding to an elastic modulus of 101 to 23 125 GPa. The flexural rigidity at the second stage was 55 to 61 N*m 2, and the 24 corresponding elastic modulus was 88 to 97 GPa.

25 Most HBR HBU SNF rods in the as

-irradiated state under static unidirectional loading 26 fractured at a location coincident with the pellet

-to-pellet interface, as validated by the 27 posttest examinations showing pellet end faces in most of the fracture surfaces.

28 Fragmentation of the pellets also occurred to a limited degree, along with cladding 29 failure. 30 The static CIRFT results indicate a significant increase in a fueled SNF rod's flexural 31 rigidity compared to a calculated response for cladding only. This applied to both as

-32 irradiated and HRT SNF rods.

33 For the HBR HBU SNF rods, the static CIRFT test results show that at bending moments 34 less than 30 N*m the flexural rigidities of the as

-irradiated rods and the HRT HR2 rod are 35 essentially the same.

36 The sample subjected to a n HRT and tested under a static bending load showed 37 reduced flexural rigidity at higher loads compared to as

-irradiated samples.

38 Nevertheless, material tested in the as

-irradiated and HRT state both had higher flexural 39 rigidity than the calculated cladding

-only response.

40 The static CIRFT test result for HR2 supports the pretest expectation (hypothesis) that 41 because the tensile bending stress in the cladding is parallel to the plane of both the 42 radial and circumferential hydrides, the presen ce of radial hydrides would not 43 2-4 significantly alter the flexural response when compared to the case where only 1 circumferential hydrides are present.

2 The methodology developed in this study calculate cladding stress and strain is 3 applicable to all cladding types, and the use of cladding

-only properties to calculate 4 cladding stress and strain is always conservative for all cladding types.

5 The HBR HBU SNF rods in the as

-irradiated state survived static unidirectional bending 6 to a maximum curvature of 2.2 to 2.5 m

-1, or a maximum m7 maximum static unidirectional bending values were bounded by the CIRFT device 8 displacement capacity. The maximum equivalent strain was 1.2 to 1.4 percent. 9 Based on the static CIRFT test results, the lower

-bound safety margin against fuel rod 10 failure during a n HAC side drop event is 2.35 assuming the side drop imparts a 50

-g 11 load to the package body (see Section 2.3.4.2). 12 The following observations can be made about the results of the dynamic testing:

13 The fatigue life of HBR HBU SNF rods in the as

-irradiated state in the cyclic tests 14 depended on the level of loading. Under loading with moments of +/-8.89 to +/-35.56 15 -namely +/-0.03 to +/-0.38 percent strain- the fatigue life ranged from 5.5 x 10 3 to 16 2.3 x 10 6 cycles. 17 The -N curve of the HBR HBU SNF rods in the as

-irradiated state can be described by 18 a power function of y = 3.839*x-0.298, where x is the number of cycles to failure, and y is 19 the strain amplitude (percent). 20 Maxima of the curvature during dynamic tests in the as

-irradiated state ranged from 21 +/-0.08 to +/-0.78 m

-1. The -N curve of the HBR HBU SNF rods can be described by a 22 power function of y = 6.864*x-0.283, where x is the number of cycles to failure, and y is the 23 -1). A fatigue limit is likely located 24 between 0.08 and 0.1 3 m-1 if it is defined at 10 7 cycles (as is typical for material fatigue 25 endurance limits).

26 The failure of HBR HBU SNF rods under cyclic loading often occurred near pellet

-to-27 pellet interfaces.

28 The samples subjected to a n HRT showed a slightly reduced lifetime compared to as

-29 irradiated samples in dynamic testing (see Section 2.3.3

). 30 The following sections provide an assessment by the NRC staff (the staff) of ORNL's CIRFT 31 data and present conclusions as to the expected structural performance of HBU SNF during dry 32 storage and transportation

. 33 2.3 Application of the Static Test Results 34 When evaluating the HAC 9-m (30-ft) drop test, as required by Title 10 of the Code of Federal 35 Regu lations (10 CFR) 71.73(c)(1), two drop orientations produce distinctly different structural 36 behaviors in the fuel rods. These orientations are the side drop and the end drop (Figure 2-2). 37 In the side drop, lateral inertia loads are applied to the fuel rods

, and bending dominates the 38 structural response. In the end drop, axial compression and the associated buckling of the fuel 39 2-5 rod dominates the structural response. For a side-drop event, the CIRFT static bending test 1 results from NUREG/CR

-7198, Revision 1 (NRC, 2017a

), can be directly applied to quantify the 2 fuel rod structural response. For the end drop, the presence of axial compression in the fuel rod 3 represents a force component that was not present in the CIRFT static bending tests. This, 4 however, does not pose a problem since the CIRFT static test results can be used to 5 conservatively quantify the effect of the fuel pellets on increasing the flexural rigidity of the rods 6 to resist buckling. 7 Figure 2-2 Schematic Diagram of End and Side Drop Accident Scenarios 8 (Revised Figure 5-168 from Patterson and Garzarolli (2015)) 9 2.3.1 Spent Fuel R od Behavior i n Bending 10 The behavior of a fuel rod in bending generally depends on three things:

(1) the type of loading

, 11 (2) the bond between the cladding and fuel

, and (3) the behavior of the pellet

-pellet interface.

12 Fundamentally

, there are two types of bending

-bending without shear and bending with shear.

13 Bending without shear is pure bending (i.e., constant moment or curvature, as exhibited in the 14 ORNL CIRFT tests) and produces no shear stress at the interface between the cladding and 15 fuel pellet. Pure bending is a special case that does not often occur in practice. What occurs 16 more often is the case of a laterally

-supported fuel rod subjected to a transverse inertia loading, 17 as in a side drop, where the rod is subjected to both bending and shear forces.

1 Although both 18 bending and shear are acting, the structural response would be expected to be different

, 19 depending on whether the cladding is bonded to the fuel pellet.

20 1 Because the fuel behaves in a brittle manner while the cladding behaves in a ductile manner, all of the bending tensile stresses will occur in the cladding.

The cladding and fuel will resist the shear forces, but for simplicity

, it can be conservatively assumed that all of the shear is resisted by the cladding. A simple calculation shows that during a side drop event

, the uniformly loaded fuel rod spanning over multiple grid spacers will have maximum tensile stresses due to bending that are more than an order of magnitude greater than the maximum tensile stresses due to shear. Therefore, bending dominates the response of the fuel rod, and this is why the CIRFT tests can accurately represent the behavior of an actual fuel rod during a side drop event.

2-6 2.3.2 Composite Behavior of a Spent Fuel R od 1 The normal explanation for the structural response of the fuel ed-rod composite system is as 2 follows. If the pellet is not bonded to the cladding, displacement compatibility is not maintained 3 at the pellet

-cladding interface

, and composite action does not occur. In this case

, the flexural 4 rigidity is given by the following

5 EI = E c I c + E p I p (E q n. 2-1) 6 That is, the flexural rigidity is equal to the sum of the individual flexural rigidities of the cladding 7 and fuel pellets, where E c and I c are the elastic modulus and moment of inertia of the cladding, 8 respectively, and E p and I p are the elastic modulus and moment of inertia of the pellet, 9 respectively. On the other hand, if the pellet is bonded to the cladding, displacement 10 compatibility is maintained at the pellet

-cladding interface and composite action occurs. In this 11 case , the flexural rigidity is calculated by transforming the pellet properties into equivalent 12 cladding properties (i.e., by multiplying the pellet moment of inertia by E p/E c). This is the same 13 technique commonly used for reinforced concrete (Winter and Nelson, 1979). As mentioned 14 above, a spent fuel rod is a composite system consisting of cladding and spent fuel. To fully 15 understand the unique behavior of this composite system, the bending behavior of a more 16 general composite beam will be discussed. Consider a composite concrete and steel I

-beam 17 where a concrete slab, rectangular in cross

-section, is poured onto the top flange of a steel I

-18 beam (Figure 2

-3). This type of composite beam is commonly found in highway bridge 19 construction. Assume the concrete and steel beam are simply supported and a concentrated 20 load is applied at mid

-span. If the concrete slab and steel beam are not bonded to each other, 21 no shear transfer takes place at the interface between the steel and concrete, and the flexural 22 rigidity (EI) is equal to the sum of the individual flexural rigidities of the concrete slab and steel 23 beam taken separately.

24 Figure 2-3 Typical Composite Construction of a Bridge 25 On the other hand, if the concrete slab and steel beam are bonded to each other, as typically 26 done using shear studs, then shear transfer takes place and the concrete slab and steel beam 27 act as a composite section. In this case

, the flexural rigidity of the composite beam will be 28 significantly greater than the sum of the individually flexural rigidities taken separately. This 29 2-7example of a concrete slab bonded to the top flange of a steel beam illustrates the behavior of a 1 composite system where the centers of gravity of each of the two components (i.e., concrete 2 slab and steel I

-beam) are not coincident.

3 For the special case where the centers of gravity of the two components are coincident, the 4 flexural rigidity of the composite section is always equal to the sum of the flexural rigidities of the 5 individual components regardless of whether the components are bonded or unbonded. The 6 following example illustrates this concept. Consider a simply supported span co mp osed of two 7 beams, each with a rectangular cross

-section 2 in. wide, and 6 in. deep (i.e., a "2 x 6"). Let the 8 2 x 6's be configured one on top of the other, where the centers of gravity (cg s) are not 9 coincident as shown in Figure 2-4a. If the beams are unbonded, the moment of inertia of the 10 section (I = bh 3/12 per beam), is equal to: 2 x 2 in. x (6 in.)3/12 = 72 in

.4. If they are bonded, 11 then the moment of inertia of the section is equal to: 2 in. x (2 x 6 in.)3/12 = 288 in.4 12 Figure 2-4 Influence of cg Position on Composite Beam Stiffness:

13 (a)cgs Are Not Coincident, (b) cgs Are Coincident 14 Now let the 2 x 6s be configured as shown in Figure 2-4b, where the cgs are aligned on the 15 same bending axis (i.e., they are "coincident

"). If they are unbonded, the moment of inertia of 16 the section is: 2 x 2 in. x (6 in.)3/12 = 72 in.4. If they are bonded I = 2 x 2 in. x (6 in.)3/12 = 72 17 in.4 Thus, w hen the cgs of the 2 x 6's are "coincident" the flexural rigidity of t he beam i s the 18 sum of the individual flexural rigidities of the 2 x 6's regardless of whether the 2 x 6s are bonded 19 or unbonded. While previously unrecognized, this is the situation with a spent fuel rod, where 20 the cladding cylindrical tube and the spent fuel cylindrical solid section have coincident c gs. 21 Thus, for a spent fuel rod, where the fuel is a homogenous solid, the flexural rigidity is given by 22 Equation 2-1, regardless of whether the fuel is bonded to the cladding. All moments of inertia 23 are taken about the neutral axis of the fuel rod.

24 Calculation of Cladding Strain 25 The objective of this section is to develop a simple methodology that uses the CIRFT static test 26 data for fully-fueled composite spent fuel rods to evaluate spent fuel rod cladding strain. T he 27 methodology presented here to determine cladding response (i.e., cladding stresses and 28 strains) is based on a set of assumptions that are consistent with those made by ORNL in its 29 2-8 presentation of CIRFT results in NUREG/CR-7198, Revision 1 (NRC, 2017a). These 1 assumptions, which are discussed in greater detail below, are based on the integrated average 2 response of the fuel rod along its gauge length.

3 Figure 2-5 Images of Cladding

-Pellet Structure in HBU SNF Rod (66.5 Gwd/MTU, 40-4 70 µm Oxide Layer, 500 Wppm H Content i n Zircaloy-4): (a) Overall Axial 5 Cross Section a nd (b) Enlarged Area (Revised Figure 33 from NUREG/CR-6 7198, Revision 1 (NRC, 2017a)) 7 The fuel rod composite system (Figure 2-5) is composed of cladding, which exhibits ductile 8 behavior, and the fuel pellet, which exhibits brittle behavior. In a spent fuel rod subject to 9 bending, where the fuel is a homogenous solid, the neutral axis is at the center of the rod cross

-10 section , provided that the brittle fuel does not crack in tension. Once the fuel cracks, the neutral 11 axis will shift toward the compression side of the cross

-section. The ORNL tests show that the 12 region of the fuel weakest in tension is at the pellet

-pellet interface. When the pellet

-pellet 13 interface cracks, the tensile stress in the cladding at the crack face

, will increase significantly.

14 On either side of the crack face the shear stress between the cladding and fuel is high an d 15 decreases parabolically with distance from the crack (Figure 2-6). The high tensile stress in the 16 cladding at the crack face also decreases parabolically with distance from the crack. Thus

, the 17 cladding tensile stresses will vary significantly along the length of the rod

they are highest at the 18 crack face and much lower away from the crack face. Even though this behavior is known to 19 occur, only the average tensile bending stress can be calculated from the static test results 20 because the measured curvature is the integrated average curvature over the measurement 21 length (gauge length) of the rod.

22 2-9 Figure 2-6 Approximate Extreme Fiber Tensile Stresses Between Pellet

-Pellet Crack 1 The LVDTs measure displacements at three locations on the test specimen. The distance 2 between the first and third probes is the gauge length of the specimen. Because the bending 3 moment is constant along the gauge length, it would be expected that several pellet

-pellet 4 interface cracks would develop within the gauge length. That being the case, the cladding 5 tensile stresses and strains along the gauge length will vary significantly. However, this 6 variation in strain along the gauge length was not, and cannot be, measured. What was 7 measured is the average curvature along the gauge length. Therefore, only the average tensile 8 strain (i.e., the smeared tensile strain) can be calculated. The average tensile strain , , along 9 the gauge length is equal to the curvature , , multiplied by the distance to the neutral axis, ymax: 10 =

• ymax (Eq. 2-2) 11 However, ymax can vary significantly along the gauge length. At a section where the fuel has not 12 cracked, ymax is equal to the outer radius, r. At a pellet

-pellet interface crack, ymax would be 13 greater than the radius but less than the diameter. However, because the measured and 14 calculated results are averages over the gauge length, a convention must be adopted for 15 calculating cladding strain and this convention must be consistently applied throughout. The 16 convention used in NUREG/CR

-7198, Revision 1 (N RC, 2017 a), and adopted in this document 17 to convert average curvature to average cladding strain, is to assume that the distance from the 18 tensile face of the cladding to the neutral axis is equal to the outside radius, r.

19 Average cladding tensile stress, , should be calculated directly from average cladding strain 20 using the following equation: 21

• E c (Eq. 2-3) 22 Use Equation 2-3 provides a consistent and compatible relationship between stress and strain.

23 2-102.3.3 Calculation of Cladding Strain Using Factored Cladding-Only Properties 1 The following discussion describes a methodology that can be easily implemented to calculate 2 the cladding tensile strain and stress and fuel rod flexural rigidity using only cladding

-only 3 properties. Section 4.2.2 of NUREG/CR

-7198, Revision 1 (N RC, 2017a), presents analyses 4 comparing the measured flexural rigidity from the CIRFT static test results to the calculated 5 flexural rigidity values using the validated cladding-only mechanical property models developed 6 by Pacific Northwest National Laboratory (PNNL) (Geelhood et al., 2008). The purpose of the 7 comparison was to investigate the effect of fuel pellets on the fuel rod's flexural rigidity and 8 cladding strain.

9 Figure 2-7 Comparison of CIRFT Static Bending Results with Calculated PNNL 11 Moment Curvature (Flexural Rigidity) Derived from Claddi ng-Only Stress

-12 Strain Curve (Reproduction of Figure 22 from NUREG/CR-7198, Revision 1 13 (N RC, 2017a)). S1 , S2 , S3, and S4 Represent the Experimental Results for 14 HBR HBU SNF As-Irradiated Specimens, HR2 Represents the Experimental 15 Results for HBR HBU SNF Hydr ide-Reoriented Specimen, and PNNL 16 Represents the Results Calculated Using the Validated Cladding-Only 17 Mechanical Property Models Developed by P NNL (From Geelhood et al., 18 2008) 19 T he CIRFT static test results plotted in Figure 2-7 show the moment

-curvatur e response of the 21 four HBR HBU SNF as-irradiated specimens S1, S2, S3, and S4 and the hydride

-reoriented 22 specimen HR2. The loading portion of the moment

-curvature response begins at 0 N

• m and 23 reaches a maximum at about 80 N

• m, at which point the specimens begin to unload. The 24 moment-curvature responses of the four HBR HBU SNF as

-irradiated specimens during loading 25 were similar up to a moment of 35 N*m. They are characterized by two distinct linear 26 responses, EI1 and EI2, followed by a nonlinear response during the loading and a linear 27 response upon unloading (EI3) (Figure 2-8). 28 2-11Also shown in Figure 2

-7 is the cladding

-only moment

-curvature loading curve constructed 1 using the PNNL cladding-only mechanical property models.

The static test results for both as

-2 irradiated and hydride

-reoriented specimens show much higher bending moment resistance 3 during loading compared to the PNNL cladding-only data. The slopes, EI1 and EI2, of the four 4 HBU fuel rods are greater than the slope of the PNNL data for the cladding-only rod. 5 Figure 2-8 Characteristic Points on Moment-Curvature Curve. A , B , C, a nd D are 7 Points o n the Curve. EI1 i s the Slope of the Loading Curve Between 0 a nd 8 A.EI2 i s the Slope of the Loading Curve Between A a nd B. EI3 i s the Slope9 of the Unloading Curve Between D and 0. The Cladding

-Only Moment

-10 Curvature Loading Curve Constructed Using the PNNL Cladding-Only 11 Mechanical Property Models is not Shown (Reproduction of Figure 21 from 12 NUREG/CR-7 198, Revision 1 (N RC, 2017a))13 Figure 2-7 also shows that at bending moments during loading less than 35 N*m, the flexural 15 rigidities of the four as

-irradiated rods, which have only circumferential hydrides, and HR2, 16 which has both circumferential and radial hydrides, are essentially the same. This result 17 supports the pretest expectation that, because the bending tensile stress in the cladding is 18 parallel to the plane of both the radial and circumferential hydrides, the presence of radial 19 hydrides would not significantly alter the flexural response from the case where only 20 circumferential hydrides are present. The results of tests currently being conducted by the U.S.

21 Department of Energy (DOE) will further confirm this hypothesis as it applies to other cladding 22 types. 23 In the CIRFT static test results for HBR HBU SNF rods shown in Figure 2

-7, no failures 25 occurred. The lower

-bound maximum moment achieved in the tests is approximately 80 N*m. 26 In addition, it is important to point out that a bending moment of 80 N*m is significantly greater 27 than the bending moment an HBR HBU SNF rod will experience during an HAC 9

-m (30-ft) side 28 drop (see Section 2.3.4.1). This means that fuel rod integrity is expected to be maintained 29 during an HAC drop scenarios, and therefore, fuel rod reconfiguration is very unlikely.

30 2-12 For the as

-irradiated HBR HBU SNF rods, Table 2

-1 shows that in the EI1 region of the 1 moment-curvature results, the average flexural rigidity is 2.66 (.e., 71.58 N*m 2/26.93 N*m 2) 2 times greater than the cladding-only case, and in the EI2 region the average flexural rigidity is 3 2.16 (i.e., 58.10 N*m 2/26.93 N*m 2) times greater than the cladding

-only case. For the 4 hydride-reoriented fuel rod, HR2, Table 2

-1 shows that in the EI1 region

, the average flexural 5 rigidity is 2.33 (i.e., 62.77 N*m 2 / 26.93 N*m 2) times greater than the cladding

-only case, and in 6 the EI2 region

, the average flexural rigidity is 1.54 (i.e., 41.52 N*m 2 / 26.93 N*m 2) times greater 7 than the cladding

-only case.

8 Table 2-1 Comparison of Average Flexural Rigidity Results Between CIRFT Static 9 Testing a nd PNNL Cladding-Only Data (From Validated Mechanical 10 Property Models in Geelhoo d e t al., 2008) 11 EI1 (N*m 2) EI2 (N*m 2) EI3 (N*m 2) As-Irradiat ed (S1, S2, S3

, and S4) 71.576 58.099 48.133 Hydride-Reoriented (HR2) 62.769 41.517 43.333 Cladding-Only (validated PNNL models) 26.933 - - Table 2-2 Characteristic Points and Quantities Based on Moment-Curvature Curves 12 (Reproduction , in Part, o f Table 4 from NUREG/CR-7198, Revision 1 13 (NRC , 2017a)) 14 Spec label EI1 2) EI2 2) EI2 2) A (m-1) B (m-1) C (m-1) D (m-1) M A M B M C M D S1 78.655 57.33 51.027 0.202 0.968 2.009 2.166 16.695 60.599 83.595 85.413 S2 73.016 60.848 52.699 0.32 1.009 2.001 2.154 20.18 62.133 85.914 87.294 S3 71.517 59.369 47.101 0.311 0.933 2.149 2.308 22.338 59.288 83.728 85.235 S4 63.117 54.849 41.704 0.503 0.862 2.329 2.507 28.54 48.244 81.656 85.02 As-irradiated Avg. 71.576 58.099 48.133 0.334 0.943 2.122 2.284 21.938 57.566 83.723 85.741 As-irradiated Std. Dev. 6.422 2.603 4.886 0.125 0.062 0.154 0.164 4.977 6.322 1.741 1.048 HR2 62.769 41.517 43.333 0.487 1.007 1.585 2.158 30.301 51.884 66.809 79.606 In developing a simplified methodology using cladding

-only mechanical properties

, the staff 15 considers it conservative to use the flexural rigidity ratio from the EI2 data. More specifically, 16 using the average minus two standard deviations of the EI2 data from Table 2-2 is 52.90 N*m 2 17 (i.e., 58.10 2 - 2 (2.60 2)), which results in an EI2 ratio of an HBU fuel rod to a 18 cladding-only rod of 1.96 (i.e., 52.90 2 / 26.93 2). The average minus two standard 19 deviations has a 98 percent exceedance probability, which means there is a 98 percent chance 20 that the actual value of the EI ratio will be greater than 1.96. To account for the effects of 21 hydride reorientation, this result is reduced by 0.713 (i.e., 1.54/2.16), which is the ratio of the 22 2-13 reoriented hydride results to the as

-irradiated resul ts that were calculated in the previous 1 paragraph. Multiplying 1.96 by 0.713 results in a factor of 1.40. However, recognizing the 2 limited test data available to calculating the 1.40 factor, the factor has been further reduced to 3 1.25 to account for the additional uncertainty associated with using limited data. Thus, for the 4 purpose of calculating lateral displacements in the simplified methodology, the flexural rigidity of 5 the HBU fuel rod is equal to the flexural rigidity of the cladding

-only rod multiplied by the factor 6 1.25: 7 (EI)HBU rod = 1.25 (EI)clad only (Eq. 2-4) 8 The curvature, , of the HBU fuel rod is given by: 9 = M/(EI)HBU rod (Eq. 2-5) 10 or: 11 = M/[1.25 * (EI)clad only] (Eq.2-6) 12 where M is the bending moment in the rod.

13 The tensile strain is given by: 14 =

• ymax (E q. 2-7) 15 where ymax is equal to the outer radius, r, of the rod, and the maximum equivalent tensile stress 16 is given by: 17 =

• E c (Eq. 2-8) 18 The methodology described above for using cladding-only properties to calculate cladding 19 strains while accounting for the increased flexural rigidity imparted by the fuel pellet can also be 20 applied to cladding alloys other than Zircaloy-4. Once CIRFT static bending results for other 21 HBU SNF rods (i.e., ZIRLO

-clad and M5

-clad rods) are obtained under planned DOE-sponsored 22 research (Hanson et al

., 2016), this methodology can be replicated to obtain a numerical factor 23 that allows for crediting the flexural rigidity of the fuel pellet in those fuel types. Until those 24 results are available, t he staff considers the use of claddin g-only mechanical properties to 25 calculate cladding stress and strain to be conservative. The staff expects that CIRFT static 26 bending results for other HBU SNF rods obtained by the DOE-sponsored research will confirm 27 this conclusion.

28 2.3.3.1 Two Alternatives for Calculating Cladding Stress and Strain During Drop 29 Accidents 30 Two alternatives can be used to calculate cladding stress and strain, and cladding flexural 31 rigidity, for the evaluation of drop accident scenarios. The first alternative is to use cladding

-32 only mechanical properties from as

-irradiated cladding (which has only circumferential hydrides

) 33 or from hydride-reoriented cladding (which would account for radial hydrides precipitated after 34 the drying process). As discussed in Section 2.3.3, the staff considers that the orientation of the 35 hydrides is not a critical consideration when evaluating the adequacy of cladding

-only 36 mechanical properties. The properties necessary to implement this alternative are derived from 37 cladding-only uniaxial tensile tests and include modulus of elasticity, yield stress, ultimate 38 2-14 tensile strength and uniform strain, and the strain at failure (i.e., the elongation strain).

1 Additional considerations for acceptable cladding

-only mechanical properties (i.e., alloy type , 2 burnup, an d temperature) may be found in either of the current standard review plans (SRPs) 3 for dry storage of SNF (NUREG-1536, Revision 1, "Standard Review Plan for Spent Fuel Dry 4 Storage Systems at a General License Facility

," issued in July 2010 (NRC, 2010) for the review 5 of applications for Certificates of Compliance under 10 CFR Part 72

and NUREG-1567, 6 "Standard Review Plan for Spent Fuel Storage Facilities," issued in March 2000 (NRC, 2000 a) 7 for the review of applications for specific licenses under 10 CFR Part 72) or transportation 8 (NUREG-1617, "Standard Review Plan for Transportation Packages for Spent Nuclear Fuel,"

9 issued in March 2000 (NRC, 2000b))

- hereafter these documents will be referred to as the 10 current SRPs for dry storage or transportation for SN F.2 11 The second alternative is to use cladding

-only mechanical properties that have been modified 12 by a numerical factor to account for the increased flexural rigidity imparted by the fuel pellet.

13 This numerical factor is obtained from static CIRFT static bending results for fully-fueled rods for 14 the particular HBU SNF cladding type and fuel type, as previously discussed. However, this 15 second alternative would be necessary only if the structural evaluation using cladding

-only 16 mechanical properties is unsatisfactory. The acceptance criteria for cladding performance 17 following dry storage and transport

-related drop accident scenarios can be found in the current 18 SRPs for dry storage and transportation of SNF, respectively.

19 2.3.4 Applicability t o Dry Storage a nd Transportation 20 Argonne National Laboratory defined the radial hydride continuity factor (RHCF) as the ratio of 21 the maximum length of continuous radial

-circumferential hydrides projected in the radial 22 direction to the cladding thickness within a 150

-m arc length (see Section 1.5.4). This metric 23 can be used to quantify the degree of reorientation induced in the hydride-reoriented specimen 24 that was static

-bend tested in the CIRFT instrument (specimen HR2). Figure 2-9 shows a 25 metallographic image of the hydride microstructure of test specimen HR1 (used for CIRFT 26 dynamic testing) after the aggressive hydride reorientation procedure used for HBR HBU SNF 27 rod specimens

.3 The HR2 specimen underwent the same radial hydride treatment (Figure 2-10) 28 as HR1, which is considered to be conservative relative to the conditions expected during drying 29 and short-term loading operations (i.e.

, bounding cladding temperature and hoop stresses, 30 multiple thermal cycling).

4 31 During the radial hydride treatment, each specimen was pressurized to induce a maximum hoop 32 stress of 140 MPa at a target temperature of 400 °C for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />, cooled at 1

°C/min to 170

°C, 33 and then heated at 1

°C/min to the hold temperature of 400

°C. This thermal cycling was 34 repeated for five cycles (a condition that HBU SNF assemblies would not experience in practice, 35 2 The current SRPs for dry storage of SNF are being consolidated into a single document, NUREG

-2215, "Standard Review Plan for Spent Fuel Dry Storage Systems and Facilities," Draft Report for Comment issued November 2017 (NRC, 2017b). Similarly, the current SRP for transportation of SNF is being revised and will be reissued as NUREG

-2216, "Standard Review Plan for Approval of Transportation Packages" (NRC, 2018). Both documents will incorporate current Interim Staff Review Guidance documents. The new SRPs will be issued for public comment and are expected to be finalized prior to final issuance of this report.

3 Section 3.4.1 of NUREG

-7198, Revisi on 1 (NRC, 2017a), presents a more detailed discussion o f the radial hydride treatment used for preparation of the Phase II specimens.

4 The cladding alloy in the HR2 test specimen (Zircaloy

-4) had a hydrogen content considered representative for cold-worked, stress

-relieved alloys (Zircaloy

-4, ZIRLO) and considered bounding for recrystallized

-annealed alloys (Zircaloy

-2, and M5).

2-15 if drying operations are performed according to the guidance in ISG

-11, Revision 3, "Cladding 1 Considerations for the Transportation and Storage of Spent Fuel," issued November 2003 2 (NRC, 20 03 a)-see prior Section 1.2 of this report) to further induce a higher fraction of radial 3 hydrides. The specimen was then furnace cooled from 170

°C to room temperature after the 4 last cycle and the pressure was released

. 5 The conservative conditions chosen for the radial hydride treatment are evidenced by the high 6 radial hydride fraction observed after metallography following testing. As Figure 2

-9 shows, the 7 conservative conditions of the radial hydride treatment induced a near-100-percent RHCF in 8 some sections of the rodlet specimen

. Since the radial hydride treatment produced the highest 9 degree of reorientation that could be anticipated, it is therefore expected to provide the most 10 limiting mechanical properties.

11 Figure 2-9 High Magnification Micrograph Showing Radial Hydrides of a HBR HBU 12 SNF Hydride-Reoriented Specimen Tested Under Phase II (Specimen HR1 13 Results Shown; Hydrogen Content 360-400 Wppm) (Reproduction of 14 Figure 35a in NUREG/CR-7198, Revision 1 (N RC , 2017a)) 15 2-16 Figure 2-10 Representative Conditions Used for Radial Hydride Treatment for 1 Preparation of HBR HBU SNF Hydride

-Reoriented Specimens Tested Under 2 Phase II.

The HBU SNF Specimen Wa s Pressurized to 140 Mpa at 400 C 3 With Five Thermal Cycles (Reproduction Of Figure 14 from NUREG/CR

-4 7198, Revision 1 (N RC, 2017a)) 5 The static test results for the hydride

-reoriented Zircaloy

-4 fuel rod (specimen HR2; Figure 2

-7) 6 show minimal difference in the flexural response compared to the as

-irradiated rods up to the 7 bending moments pertinent to a 9

-m (30-ft) drop accident (i.e., bending moments below 35 N

• m 8 - see Section 2.3.4.2 for pertinent calculation). More importantly, the flexural rigidity of the 9 hydride-reoriented specimen is still markedly higher than the calculated cladding-only response 10 according to validated PNNL mechanical property models. The major difference between the 11 response of the hydride-reoriented HR2 specimen and the as

-irradiated rods is the slightly lower 12 flexural resistance of HR2 at higher loads. The slightly lower flexural resistance at higher loads 13 may be the result of the higher density of hydrides in HR2 or the greater extent to which 14 debonding occurred between the cladding and pellet away from the pellet

-to-pellet crack 15 interface.

However, those loads would not be expected during transportation or dry storage 16 operations.

17 The static test results for the hydride

-reoriented HR2 and the as

-irradiated HBR HBU SNF 18 Zircaloy-4-clad fuel rods support the staff's conclusion that the use of cladding

-only mechanical 19 properties is adequate for the structural evaluation of HAC and NCT drop events. Further, t he 20 HAC drop events required for transportation packages apply inertia loads to the fuel rods that 21 bound the design basis storage drops (e.g., drops during transfer operations and no n 22 mechanistic tip over). Therefore, this conclusion based on the CIRFT static test results of 23 Zircaloy-4 can be applied to both transportation and storag

e. 24 The cladding strains that control the static response of an intact fuel rod are the high tensile 25 strains at the face of the crack at the pellet

-pellet interface.

If a pinhole or hairline crack were to 26 be present at this location

, it could have an effect on the static test results because of the strain 27 concentrations they may create. However, the staff considers the probability that a pinhole or 28 hairline crack is at the pellet

-pellet crack face simulateneously longitudinally and 29 2-17 circumferentially to be low. Therefore, it is reasonable that the CIRFT static test results for 1 intact fuel rods can also be applied to undamaged fuel with pinholes or hairline cracks.

2 Since the hydride

-reoriented Zircaloy

-4 fuel rod had a nearly 100 percent RHCF, the staff 3 considers that the same response should be observed by all modern commercial cladding alloy 4 types that may experience hydride reorientation (i.e., Zircaloy-2, ZIRLO and M5). The staff has 5 also reviewed proprietary and non

-proprietary data on end

-of-life rod internal pressures for fuel 6 rods with boron

-based integral fuel burnable absorbers (see Section 1

.5.3) and considers that 7 these rods are also reasonably bound by the maximum rod internal pressure used in the CIRFT 8 radial hydride treatment (i.e., 140 MPa). The staff's expectation is that future DOE

-sponsored 9 CIRFT static testing conducted on other cladding alloy types will provide confirmation of this 10 conclusion (Hanson et al

., 2016). 11 2.3.4.1 Use o f Static Test Results to Evaluate Safety Margins in an HAC Side Drop 12 Event 13 The CIRFT static test results can be used to determine a lower bound safety margin against fuel 14 rod failure during a n HAC side drop event. The safety margin is calculated by dividing the load 15 (or moment) at rod failure by the maximum applied load (or moment) occurring during the side 16 drop event.

17 Figure 2-7 shows that static testing of the HBR HBU SNF rods did not result in rod failures. The 18 lower bound maximum moment achieved in the tests is approximately 80 Nm. Based on the 19 slope of the curves at 80 Nm, it is reasonable to assume that rod failure probably occurs at a 20 moment at or below 100 Nm. Therefore, using 80 Nm provides a conservative basis for 21 calculating safety margin. To quantify the safety margin it is necessary to know the bending 22 moment in the fuel rod as a function of the g

-load acting on the rod due to a side drop event.

23 Each fuel rod in the fuel assembly is supported by grid spacers at multiple locations along the 24 rod. Therefore, for the purpose of calculating the maximum bending moment, the rod can be 25 idealized as a uniformly loaded continuous beam.

26 Relationship Between Applied G-Load and Bending Moment 27 For the purpose of evaluating a safety margin, two different fuel rods are initially considered.

28 The first is a fuel rod from a PWR 15 x 15 fuel assembly, and the second is a n HBR fuel rod 29 that was tested by ORNL in the CIRFT testing device and reported in NUREG/CR

-7198 , 30 Revision 1 (N RC, 2017 a). 31 The properties of the PWR 15 x 15 rod (Table 2-3) are taken from NUREG

-1864, "A Pilot 32 Probabilistic Risk Assessment of a Dry Cask Storage System at a Nuclear Power Plant,"

33 Appendix C, Table C.1, issued March 2007 (NRC, 2007 a). 34 Table 2-3 Properties of PWR 15 x 15 SNF R od 35 Total fuel r od weight 7.011 lb Fuel length 154 in. Number of grid spacers 8 Rod length between g rid spacers (l) 20.5 in. Uniform applied load (w = 7.011 lb / 154 in

.) 0.0455 lb/in

.

2-18 The maximum moment in a uniformly

-loaded continuous beam can be approximated by the 1 maximum moment in a uniformly loaded three

-span continuous beam as shown in Eqn. 2-9: 2 Mmax = 0.100

• w

• l 2 (Eqn. 2-9) 3 i.e., Mmax = (0.100)(0.0455 lb/in.)(20.5 in.)2 = 1.91 lb*in. = 0.2 16 N*m 4 This is the moment resulting from a 1 g

-loading. The g

-load necessary to produce a moment of 5 1 N*m = 1 g / 0.2 16 N*m = 4.63 g / N*m. 6 For the HBR HBU SNF rod, the weight per unit length is calculated from the weight density of 7 fuel and the weight density of cladding, which can be determined from the information in 8 NUREG-1864 , Table C.1 (NRC, 2007 a) for a BWR 7 x 7 fuel rod.

9 Fuel density = 0.34 lb / in.3 10 (i.e., 9.60 lb / [()(0.25)2(144)] = 0.34) 11 Cladding density = 0.234 lb / in

.3 12 (i.e., 1.98 /

[()(0.535)(0.035)(144)

] = 0.234) 13 The diameter (outer, inner) and thickness of the cladding of a n HBR HBU SNF rod as given in 14 NUREG/CR-7198, Revision 1 (N RC, 2017 a) are: 15 Outer diamete r = 10.743 mm = 0

.423 in. 16 Cladding thickness

= 0.748 mm = 0.0294 in

. 17 Inner diameter

= 0.364 in

. 18 From the HBR HBU SNF rod cross-sectional dimensions and the fuel and cladding densities 19 calculated using the data for the BWR 7 x 7 fuel rods, the fuel and cladding weight per unit 20 length can be calculated as follows:

21 HBR fuel weight = 0.0354 lb / in. 22 HBR cladding weight = 0.0085 lb

/ in. 23 w = 0.0354 + 0.0085 = 0.0439 lb / in. 24 l = distance between HBR SNF assembly grid spacers = 26.2 in

. 25 Mmax = (0.100)(0.0439)(26.

2)2 = 3.01 lb*in = 0.340 N*m 26 This is the moment resulting from a 1 g

-loading. The g

-load necessary to produce a moment of 27 1 N*m = 1 g / 0.340 N*m = 2.94 g / N

• m. 28 This example illustrates the fact that the static transverse g

-load necessary to produce a 29 bending moment of 1 N*m in a fuel rod supported by multiple grid spacers varies from rod to 30 rod. For the two rods in this example, the static transverse g

-load required to produce a 31 2-19 bending moment of 1 N*m varied from 2.9 to 5 g depending on the rod cross sectional 1 dimensions and assembly geometry.

2 2.3.4.2 Dynamic Response of a Fuel Rod 3 During a HAC 9

-m (30-ft) side drop of a transportation package with impact limiters, the cask 4 body will typically experience inertia loads on the order of 50 g. However, the fuel rod is flexible, 5 as are the intervening components that support the rod between the cask body and the rod.

6 Therefore, the rigid body deceleration of the cask body will be amplified during a side drop event 7 by the flexibility of the rod and intervening components, resulting in a g

-load in the rod that is 8 higher than the g

-load acting on the cask body. This increase in g

-load is expressed by a 9 dynamic load factor (DLF), which is the ratio of the deflection due to a dynamically applied load 10 to the deflection that would have resulted from the static application of the load.

The DLF will 11 depend on the rod's natural frequency, the duration of the loading, and the shape of the load 12 time history.

13 Since natural frequency, load duration and load time history shape all depend on the physical 14 characteristics of the fuel assembly, the rod and the cask, including impact limiters, a 15 conservative approach will be used to calculate safety margin by using a maximum DLF of 2.0 16 (Biggs, 1964). 17 Thus, the statically equivalent g

-load the fuel rod is subjected to is 18 (DLF) * (50 g) = 2.0 * (50 g) = 100 g 19 which produces a bending moment in the rod of 20 100 g / (2.94 g/N*m) = 34.0 N*m 21 The safety margin (SM) against fuel rod failure during a side drop event is then 22 SM = (80 N*m)/(34.0 N*m) = 2.35 23 2.3.4.3 Seismic Response o f a Fuel Rod 24 The seismic response of a fuel rod can be determined using a variety of structural models.

25 These range from simple idealized models, for which hand calculation methods could be used, 26 to very detailed finite element models. The seismic loads can be applied to these models using 27 either the response spectrum method or a time history analysis method. However, regardless 28 of whether the fuel rod is in a DSS or transportation package, seismic loads will not dominate 29 fuel rod response, because the g

-loads produced by a seismic event are not large enough. In 30 storage the g

-loads applied to the fuel are dominated by the non

-mechanistic tipover event and 31 in a transportation package the g-loads applied to the fuel rod are dominated by the HAC. Both 32 of these events produce g

-loads on the fuel rod that are approximately an order of magnitude 33 larger that the g

-loads produced by a seismic event. In addition, these two events do not occur 34 coincidently with a seismic event and therefore the seismic event does not add to either of these 35 two events.

36 2-20 2.4 Application of Fatigue Test Results 1 2.4.1 Lower Bound Fatigue S-N Curves 2 Fatigue strength data are commonly presented in the form of an S

-N curve, where S is a 3 strength parameter, such as stress or strain, and N denotes the number of cycles to failure at a 4 specific value of the strength parameter. The objective of this section is to develop a lower 5 bound fatigue S

-N curve, that envelopes the HBR HBU Zircaloy

-4 fuel rod fatigue data and 6 includes both as-irradiated rods and rods with reoriented hydrides.

7 Table 2-4 presents the fatigue test data for the HBR HBU fuel rods. In Figure 2-11 , half of the 8 cladding strain range (which isin Table 2-4 and the maximum strain (//max) are plotte d 9 against the number of cycles required to produce cladding failure at a particular strain 10 amplitude.

The strain range is the average of the strains caused by positive and negative 11 bending moments

, which produce different values of curvature and hence strain. The maximum 12 strain is the maximum of the se two strains.

13 Table 2-4 Summary o f CIRFT Dynamic Test Results for A s-Irradiated and Hydride

-14 Reoriented HBR HBU SNF (Reproduction of Table 6 i n NUREG/CR-7198, 15 Revision 1 (NRC, 2017a)) 16 Spec label Seg. ID Load amp. (N) Moment amp. Number of cycles Failure a (m-1) max m1) a (MPa) a (percent) max (percent) D0 605D1F 250 24.068 2.50E+04 Yes 0.439 0.444 206.109 0.236 0.239 D1 607C4B 150 14.107 1.10E+05 Yes 0.215 0.24 117.26 0.117 0.13 D2 608C4B 50 4.207 6.40E+06 No 0.046 0.067 35.496 0.025 0.036 D3 605C10A 100 9.17 1.00E+06 Yes 0.125 0.171 77.938 0.067 0.092 D4 605D1C 75 6.726 1.10E+07 No 0.089 0.12 57.596 0.048 0.065 D5 605D1B 90 8.201 2.30E+06 Yes 0.114 0.123 69.706 0.061 0.066 D6 609C4 125 11.624 2.50E+05 Yes 0.205 0.218 99.546 0.11 0.117 D7 609C3 200 18.923 6.50E+04 Yes 0.351 0.37 160.835 0.189 0.199 D8 606C3E 87.5 7.743 1.28E+07 No 0.107 0.118 66.309 0.057 0.063 D9 609C7 350 33.667 7.10E+03 Y es 0.576 0.624 288.308 0.31 0.335 D10 606C3A 125 11.552 1.80E+05 Yes 0.174 0.213 98.185 0.094 0.115 D11 607C4A 300 29.021 5.50E+03 Yes 0.469 0.564 241.223 0.254 0.306 D12 608C4A 110 9.986 3.86E+05 Yes 0.144 0.171 83.617 0.078 0.092 D13 606B3E 135 12.55 1 1.29E+05 Yes 0.151 0.199 106.677 0.081 0.107 D14 606B3D 87.5 7.842 2.74E+05 Yes 0.112 0.135 66.652 0.06 0.073 D15 606B3C 75 6.639 2.24E+07 No 0.087 0.125 56.426 0.047 0.067 HR1 607D4C 150 15.152 4.19E+04 Yes 0.424 0.433 128.788 0.228 0.233 HR3 608D4A 100 8.982 2.44E+05 Yes 0.219 0.233 76.342 0.118 0.125 HR4 608D4C 160 14.759 5.47E+04 Yes 0.323 0.344 125.449 0.174 0.185 2-21 Figure 2-11 Plots of Half of the Cladding Strain Range (/2) a nd the Maximum Stra in 1 (//Max) a s a Function of Number of Cycles to Failure.

Markers with Arrows 2 Indicate that the Tests Were Stopped Without Failure. (Reproduction of 3 Figure 31b In NUREG/CR 7198, Revision 1 (NRC, 2017a)) 4 The lower bound enveloping S

-N curve for the HBR HBU SNF rods is composed of three 5 straight line segments when plotted on a linear

-log scale. To account for uncertainty with 6 respect to future test results and the influence of higher test temperatures, the equivalent strain 7 amplitude of all segments has been reduced by a factor of 0.9. The 0.9 is justified to account 8 for uncertainty with respect to future test results particularly at higher temperatures.

Each 9 segment's beginning and end point labels from Table 2-4 coordinates (equivalent strain 10 amplitude percent, number of cycles to failure) are given in Table 2-5 and plotted in Figure 2

-12. 11 Table 2-5 Coordinates for Lower-Bound Enveloping S-N Curve for the HBR HBU SNF 12 Rods (Equivalent Strain Amplitude Percent, Number of Cycles to Failure) 13 Segment Beginning Point End Point 1 (D11 to D13)

(0.275, 5.50E+3)

(0.096, 1.29E+5) 2 (D13 to D14)

(0.096, 1.29E+5)

(0.066, 2.74E+5) 3 (D14 to D15)

(0.066, 2.74E+5)

(0.06, 2.4E+)

2-22 Figure 2-12 CIRFT Dymanic (Fatigue) Test Results for As

-Irradiated and Hydride

-1 Reoriented H.B. Robinson Zircaloy

-4 HBU Fuel Rods. The Calculated 2 Lower-Bound Fatigue Endurance Curve is also Shown 3 Fatigue data for reoriented cladding alloys other than Zircaloy

-4 (i.e., Zircaloy

-2, ZIRLO and M5) 4 are not yet available. However, the staff believes the methodology described above for 5 developing a lower

-bound fatigue curve can be used to construct a lower

-bound fatigue curve 6 for other cladding alloys once the as

-irradiated fatigue data become available. The fatigue data 7 plotted in Figure 2

-11 show that at the same number of cycles all of the Zircaloy

-4 fuel rods with 8 reoriented hydrides failed at nearly the same strains as the as

-irradiated Zircaloy-4 fuel rods. 9 Rod specimen D2, which did not fail, was tested at a very low moment amplitude resulting in a 10 very low maximum strain amplitude. The test was also terminated prematurely at 6.4 x 10 6 11 cycles. Based on the results for the other test specimens that did not fail, it would be expected 12 that specimen D2 would not have failed until 1 x 10 8 cycles or beyond. Therefore, rod specimen 13 D2 is not included in the development of the lower bound curve since it would have 14 inappropriately skewed the results. Therefore, the staff considers that a lower

-bound fatigue 15 curve developed from as

-irradiated data for other cladding alloys is adequate for assessing the 16 fatigue life of alloys with reoriented hydrides. Additional fatigue data for hydride

-reoriented 17 specimens for other cladding alloys to be obtained under DOE

-sponsored research are 18 expected to confirm these expectations.

19 2.4.2 Fatigue Cumulative Damage Model 20 During NCT if a fuel rod were to vibrate at a constant strain amplitude, all that would be 21 necessary to predict the fatigue life of the rod is the S

-N curve. However, fuel rod vibration 22 during NCT is expected to have a series of many cycles encompassing a range of strain 23 amplitudes and with each cycle, damage to the fuel rod cladding is continuously accumulating.

24 A fatigue damage model can be used to express how damage from these cycles accumulates.

25 To date, more than 50 fatigue damage models have been proposed, but unfortunately none of 26 these models enjoys universal acceptance, and the applicability of each model varies from case 27 to case. Unlike the aerospace industry, which has conducted extensive research on the 28 2-23 accumulation of fatigue damage to materials, such as steel, aluminum, and titanium, no 1 research has been conducted on fatigue damage to HBU spent fuel cladding. Nevertheless, for 2 many metals, the simple linear damage rule developed by Miner (Gaylord and Gaylord, 1979) 3 appears to provide a simple and reasonably reliable prediction of fatigue behavior under random 4 loadings, and therefore, will be used to evaluate fatigue damage accumulation in HBU SNF rods 5 during NCT.

6 For failure, the linear damage rule is, the following:

7 i = n i/N i = n 1/N 1 + n 2/N 2 + n 3/N 3 + ... = 1 (Eqn 2-9) 8 where: 9 n i = number of strain cycles at strain level i 10 N i = number of strain cycles to produce failure at i. 11 To apply this simple linear damage rule it is assumed that the NCT loading history can be 12 reduced to a series of different strain levels where the number of cycles associated with each 13 strain level, i, is, n.. To account for uncertainty in using a simple linear damage rule to describe 14 the accumulated fatigue damage in HBU fuel, the right side of the above equation should be set 15 equal to 0.7. This value is considered an approximate lower bound for the uncertainty in Miner's 16 damage model (Hashin, 1979).

17 2.4.3 Applicability t o Storage a nd Transportation 18 The CIRFT fatigue tests were conducted under conditions that produced a uniform bending 19 moment in the fuel rod. Thus, these results apply only to loading conditions that produce 20 longitudinal bending stresses in the cladding of the fuel. Such loading conditions occur when 21 fuel rods vibrate during NCT.

Fluctuating loads can also occur during storage when the 22 cladding experiences thermal cycles because of daily and seasonal fluctuations in ambient 23 temperature. These thermal cycles will induce cyclic stresses on the cladding due to changes in 24 fission and decay gas pressure

, which will result in fluctuation s in cladding hoop stresses. As 25 explained above, however, the fatigue test results apply only to loading conditions that produce 26 longitudinal bending stresses in the cladding of the fuel. The fatigue test results are not 27 applicable to loading conditions that produce fluctuations in hoop stress. Therefore, the fatigue 28 test results cannot be applied to thermal fatigue during storage.

29 In the CIRFT static and fatigue tests the fuel rods were subjected to a constant bending moment 30 which resulted in a longitudinal bending stress in the cladding. However, in an actual spent fuel 31 rod there is internal gas pressure, which creates hoop stress on the order of 100 MPa

- see 32 Section 1.5.3.3. The presence of the hoop stresses creates a non

-proportional biaxial stress 33 state in the cladding. The stress state is non

-proportional because the hoop stress remains 34 constant while the longitudinal bending stress fluctuates. Recent research on the effect of 35 proportional biaxial stress fields on fatigue crack growth show s no significant effect of the biaxial 36 stress field on fatigue crack propagation behavior (Pickard , 2015). It is expected that the same 37 result would also hold for non-proportional biaxial stress fields. Based on these results, the staff 38 considers that the presence of a biaxial stress field in a spent fuel rod does not need to be 39 considered Therefore, only the longitudinal bending stresses in the cladding need to be 40 considered when using the ORNL static and fatigue test data.

41 2-24 During storage or transportation, it is possible that a seismic event could occur. Typically the 1 strong motion duration of a seismic event is approximately 10 seconds. A fuel rod generally 2 responds to seismic input in the 10 to 30 h ertz (Hz) frequency range. This means that the 3 number of fatigue cycles associated with a seismic event would be no more tha n about 300 4 cycles (10 seconds x 30 Hz = 300 cycles). In addition, it is expected that the seismic load 5 applied to the rod would be less than 10

-g. Based on the results summarized at the end of 6 Section 2.3.4.1, a 10

-g load would produce a bending moment in the rod of about 3.5 N*m. 7 From Table 2

-4, a bending moment of 3.5 N*m would result in a maximum cladding strain of 8 about 0.03%. From an event that produced 300 bending cycles at a maximum strain of 9 0.03%, Figures 2

-11 and 2-12 show that virtually no fatigue damage would be expected.

10 Therefore, seismic events during storage or transportation are not expected to compromise 11 the fuel integrity.12 3-1 3 DRY STORAGE OF HIGH BURNUP SPENT NUCLEAR FUEL 1 3.1 Introduction 2 The U.S. Nuclear Regulatory Commission (NRC) staff (the staff) has developed example 3 licensing and certification approaches for dry storage of high burnup (HBU) spent nuclear fuel 4 (SNF). Applicants may use these approaches to provide reasonable assurance of compliance 5 with Title 10 of the Code of Federal Regulations (10 CFR) Part 72, "Licensing Requirements for 6 the Independent Storage of Spent Nuclear Fuel, High

-Level Radioactive Waste, and Reactor 7 Related Greater Than Class C Waste,"

during normal, off

-normal and accident conditions of 8 storage. The staff developed these example approaches according to the conclusions of the 9 engineering assessment in Chapter 2. Figure 3-1 provides a hig h-leve diagram of these 10 approaches, which vary based on (1) the condition of the fuel (undamaged or damaged), and 11 (2) the length of time the fuel has been in dry storage. Section 3.2.2. discusses considerations 12 for additional analyses expected for non

-leaktight dry storage system (DSS) designs. An 13 applicant may consider and demonstrate other approaches that may be acceptable. 14 As required by 10 CFR 72.24(b) and 10 CFR 72.236(a), an application for a specific license for 15 an independent spent fuel storage installation (ISFSI) or an application for a Certificate of 16 Compliance (CoC) for a DSS design, respectively, should identify the allowable SNF contents 17 and condition of the assembly and rods per the design bases

. The allowable cladding condition 18 for the S NF contents is generally defined in the Technical Specifications of the specific license 19 (10 CFR 72.44(c)) or CoC (10 CFR 72.

236 (a)), and the nomenclature may vary between 20 different DSS designs. For example, the terms "intact" and "undamaged" have both be en 21 historically used to describe cladding without any known gross cladding breaches. In 22 accordance with 10 CFR 72.212(a)(1) and 10 CFR 72.212(b)(3), users of DSSs (general 23 licensees) are to comply with the Technical Specifications of the CoC by selecting and loading 24 the appropriate fuel, and are to maintain records that reasonably demonstrate that loaded fuel 25 was adequately selected, in accordance with their approved site procedures and Quality 26 Assurance Program. 27 Interim Staff Guidance (ISG)-1 , Revision 2, "Classifying the Condition of Spent Nuclear Fuel for 28 Interim Storage and Transportation Based on Function," issued in May 2007 (NRC, 2007b

), 29 provides guidance for developing the technical basis supporting the conclusion that the SNF 30 (both rods and assembly) to be loaded in a DSS are intact or undamaged.

1 This would include 31 considering whether the material properties, and possibly the configuration, of the SNF 32 assemblies can be altered during the requested dry storage period. If th e alteration is 33 significant enough to prevent the fuel or assembly from performing its intended functions, then 34 the fuel assembly should be classified as damaged.

35 Damaged SNF is generally defined in terms of the characteristics needed to perform functions 36 to ensure compliance with fuel-specific and DSS-related regulations. A fuel

-specific regulation 37 defines a characteristic or performance requirement of the SNF assembly. Examples of such 38 regulations include 10 CFR 72.122(h)(1) and 10 CFR 72.122(l). A DSS-related regulation 39 defines a performance requirement placed on the fuel so that the DSS can meet its regulatory 40 requirements. Examples of such regulations include 10 CFR 72.122(b) and 10 CFR 72.124(a).

41 1 The current revisions of all ISG documents will be rolled into revised standard review plans (SRPs) for dry storage and transportation, as appropriate, and will then be removed from the public domain. The revised SRPs will be issued for public comment prior to being finalized.

3-2The glossary in this report provides the staff's definitions of intact, undamaged, and damaged 1 fuel. For additional information, refer to the current standard review plans (SRPs) for dry 2 storage of SNF (NUREG

-1536, Revision 1, "Standard Review Plan for Spent Fuel Dry Storage 3 Systems at a General License Facility

," issued in July 2010 (NRC, 2010) for the review of 4 applications for Certificates of Compliance under 10 CFR Part 72, and NUREG

-1567, "Standard 5 Review Plan for Spent Fuel Storage Facilities," issued in March 2000 (NRC, 2000a) for the 6 review of applications for specific licenses under 10 CFR Part 72) - hereafter, these documents 7 will be referred to as the current SRPs for dry storage SNF.

2 8 2 The current SRPs for dry storage of SNF are being consolidated into a single document, NUREG

-2215, "Standard Review Plan for Spent Fuel Dry Storage Systems and Facilities," Draft Report for Comment issued November 2017 (NRC, 2017b), which will incorporate current Interim Staff Review Guidance documents. NUREG-2215 has been issued for public comment and is expected to be finalized prior to final issuance of this report.

3-3 Figure 3-1 Example Licensing and Certification Approaches for Dry Storage of High Burnup Spent Nuclear Fuel

3-4 Consistent with the guidance in (ISG)-1 , Revision 2 (NRC, 2007b

), HBU SNF assemblies with 1 any of the following characteristics, as identified during the fuel selection process, are generally 2 classified as damaged unless an adequate justification is provided for not doing so

3 There is visible deformation of the rods in the HBU SNF assembly. This does not refer 4 to the uniform bowing that occurs in the reactor; instead, this refers to bowing that 5 significantly opens up the lattice spacing.

6 Individual fuel rods are missing from the assembly. The assembly may be classified as 7 intact or undamaged if the missing rod(s) do not adversely affect the structural 8 performance of the assembly, or radiological and criticality safety (e.g., there are no 9 significant changes to rod pitch). Alternatively, the assembly may be classified as intact 10 or undamaged if a dummy rod that displaces a volume equal to, or greater than, the 11 original fuel rod is placed in the empty rod location.

12 The HBU SNF assembly has missing, displaced, or damaged structural components 13 such that either:

14 - Radiological and/or criticality safety is adversely affected (e.g., significant change 15 in rod pitch), 16 - The structural performance of the assembly may be compromised during normal, 17 off-normal , and accident conditions of storage, or 18 - The assembly cannot be handled by normal means (i.e., crane and grapple), if 19 the design bas es relies on ready retrieval of individual fuel assemblies.

20 Reactor operating records or fuel classification records indicate that the HBU SNF 21 assembly contains fuel rods with gross rupture

. 22 The HBU SNF assembly is no longer in the form of an intact fuel bundle (e.g., consists 23 of, or contains, debris such as loose fuel pellets or rod segments).

24 Defects such as dents in rods, bent or missing structural members, small cracks in structural 25 members, and missing rods do not necessarily render an assembly as damaged, if the intended 26 functions of the assembly are maintained

i.e., the performance of the assembly does not 27 compromise the ability to meet fuel

-specific and DSS-related regulations.

28 3.2 Uncanned Fuel (Intact and Undamaged Fuel) 29 Undamaged HBU SNF can be stored in the D SS without the need for a separate fuel can (i.e.

, a 30 separate metal enclosure sized to confine damaged fuel particulates) to maintain a known 31 configuration inside the DSS confinement cavity. This fuel includes rods that are either intact 32 (i.e., no breaches of any kind) or that contain small cladding defects (i.e.

, pinholes or hairline 33 cracks) that may permit the release of gas from the interior of the fuel rod. Cladding with gross 34 ruptures that may permit the release of fuel particulate s may not be considered undamaged. The 35 configuration of undamaged HBU SNF may be demonstrated to be maintained if loading and 36 transport operations are designed to prevent and/or mitigate degradation of the cladding and 37 other assembly components, as discussed in ISG-22 , "Potential Rod Splitting Due to Exposure to 38 an Oxidizing Atmosphere during Short

-Term Cask Loading Operations in LWR or Other Uranium 39 Oxide Based Fuel," issued May 2006 (NRC, 2006).

40 3-5 Following the approaches delineated in Figure 3-1, an application for dry storage of undamaged 1 HBU SNF would include a structural evaluation of the fuel rods under design

-bases drop 2 accident scenarios. The evaluation serves to demonstrate that the uncanned fuel remains in a 3 known configuration after a drop accident scenario.

4 Two alternatives may be used to calculate cladding stress and strain, and cladding flexural 5 rigidity, for the aforementioned evaluation of drop accident scenarios. The first alternative, 6 shown in Figure 3-2, is to use cladding

-only mechanical properties from as

-irradiated cladding 7 (i.e., cladding with circumferential hydrides, primarily), or hydride

-reoriented cladding (i.e., 8 cladding that account s for radial hydrides precipitated after the drying process).

9 Figure 3-2 First Approach for Evaluating Design

-Bases Drop Accidents During Dry 10 Storage 11 As discussed in Section 2.3.3, the staff considers the orientation of the hydrides not to becritical 12 when evaluating the adequacy of cladding

-only mechanical properties.

Therefore, the properties 13 necessary to implement this first alternative may be derived from cladding

-only uniaxial tensile 14 tests and include modulus of elasticity, yield stress, ultimate tensile strength and uniform strain, 15 and the strain at failure (i.e., the elongation strain). Refer to the current SRPs for dry storage of 16 SNF for additional considerations for acceptable cladding

-only mechanical properties (i.e., alloy 17 3-6 type, burnup, and temperature) and the acceptance criteria for cladding performance during dry 1 storage operations.

2 A second alternative , shown in Figure 3.3, is to use cladding

-only mechanical properties that 3 have been modified by a numerical factor to account for the increased flexural rigidity imparted 4 by the fuel pellet. This numerical factor is obtained from static test data from the cyclic 5 integrated reversible

-bending fatigue tester (CIRFT) for fully-fueled rods for the particular 6 cladding type and fuel type (see Section 2.3.3

). However, this second alternative would be 7 necessary only if the structural evaluation using cladding

-only mechanical properties is 8 unsatisfactory. Refer to the current SRP for dry storage of SNF for acceptance criteria on 9 cladding performance during dry storage operations

. 10 Figure 3-3 Second Approach for Evaluation of Design-Bases Drop Accidents During 11 Dry Storage 12 3.2.1 Leaktight Confinement 13 Consistent with the guidance in the current SRPs for dry storage of SNF, an application for a 14 DSS for HBU SNF is expected to define the maximum allowable leakage rate for the entire 15 confinement boundary. The maximum allowable leak age rate is based on the quantity of 16 radionuclides available for release and is evaluated to meet the confinement requirements for 17 3-7 maintaining an inert atmosphere within the DSS confinement cavity and compliance with the 1 regulatory limits of 10 CFR 72.104, "Criteria for Radioactive Materials in Effluents and Direct 2 Radiation from an ISFSI or MRS,"

and 10 CFR 72.106, "Controlled Area of an ISFSI or MRS."

3 L eak age rate testing is performed on the entire confinement boundary (over the course of 4 fabrication and loading) and ensures that the package can maintain a leak rate below the 5 maximum allowable leakage rate per ANSI N14.5 (2014). 6 If the entire DSS confinement boundary, including its closure lid, is designed and tested to be 7 "leaktight" as defined in American National Standards Institute (ANSI) N14.5

- 2014, "American 8 National Standard for Radioactive Materials

-Leakage Tests on Packages for Shipment" and the 9 current SRPs for dry storage of SNF, then the application is not expected to include additional 10 dose calculations based on the allowable leakage rate that demonstrate compliance with the 11 regulatory limits of 10 CFR 72.104(a) and 10 CFR 72.106(b). In addition , the structural analysis 12 of the package is to demonstrate that the confinement boundary will not fail under the postulated 13 drop scenarios and that the confinement boundary will remain leaktight under all conditions of 14 storage. Refer to the current SRPs for dry storage of SNF for additional guidance on 15 demonstrating compliance with the leaktight criterion.

16 3.2.2 Non-Leaktight Confinement 17 For those DSS designs not tested to a "leaktight" confinement criterion, the application is 18 expected to include dose calculations based on the allowable leakage rate to demonstrate 19 compliance with the regulatory limits of 10 CFR 72.104(a) and 10 CFR 72.106(b). L eak age rate 20 testing is performed on the entire confinement boundary (over the course of fabrication and 21 loading) and ensures that the package can maintain a leak rate below the maximum allowable 22 leakage rate per ANSI N14.5 (2014). 23 To determine the dose rate for the confinement boundary, an application for a non

-leaktight 24 DSS is expected to provide a technical basis for the assumed bounding HBU fuel failure rates 25 for normal, off

-normal, and accident conditions of storage. If an application is not able to 26 provide and justify its bounding fuel failure rates, then the fuel failure rates below can be 27 assumed as bounding values for normal, off

-normal, and accident conditions of storage

28 Normal conditions of storage: 1 percent 29 Off-normal conditions of storage: 10 percent 30 Accident conditions of storage: 100 percent 31 Bounding Release Fractions for High Burnup Spent Nuclear F uel 32 HBU SNF fuel has different characteristics than low burnup (LBU) SNF with respect to cladding 33 oxide thickness, hydride content, radionuclide inventory and distribution, heat load, fuel pellet 34 grain size, fuel pellet fragmentation, fuel pellet expansion and fission gas release to the rod 35 plenum [See Appendix C.5 to NUREG/CR-7203, "A Quantitative Impact Assessment of 36 Hypothetical Spent Fuel Reconfiguration in Spent Fuel Storage Casks and Transportation 37 Packages," issued September 2015 (NRC, 2015) for additional details on HBU SNF]. These 38 characteristics may affect the mechanisms by which the fuel can breach and the amount of fuel 39 that can be released from failed fuel rods.

Hence, the staff evaluated open literature on HBU 40 fuel rod failure rates and release fractions of Chalk River unknown deposits (CRUD), fission 41 gases, volatiles, and fuel fines to assist in the review of applications for non

-leaktight 42 3-8 confinement boundaries.

Table 3-1 provides release fractions that may be considered 1 reasonably bounding for HBU SNF.

If the release fractions are not used, justification of the 2 proposed release fractions of the source terms is expected to include an adequate description 3 of burnup for the test specimen, number of tests, collection method for quantification of 4 respirable release fractions, test specimen pressure at the time of fracture, and source 5 collection system (sophisticated enough to gather the bounding respirable release fractions

). 6 Table 3-1 Fractions of Radioactive Materials Available for Release from HBU SNF 7 Under Conditions of Dry Storage (for both Pressurized Water Reactor a nd 8 Boiling Water Reactor Fuels) 9 Variable Normal Conditions Off-Normal Conditions Accident-Fire Conditions Accident-Impact Condit ions Fraction of Fuel Rods Assumed to Fail 0.01 0.1 1.0 1.0 Fraction of Fission Gases Released Due to a Cladding Breach 0.15 0.15 0.15 0.35 Fraction of Volatiles Released Due to a Cladding Breach 3 x 10-5 3 x 10-5 3 x 10-5 3 x 10-5 Mass Fraction of Fue l Released as Fines Due to a Cladding Breach 3 x 10-5 3 x 10-5 3 x 10-3 3 x 10-5 Fraction of CRUD Spalling Off Cladding 0.15 0.15 1.0 1.0 CRUD 10 The average CRUD thickness in HBU SNF cladding has been estimated to be similar to that 11 observed on LBU SNF cladding. A review of data in the literature (NRC, 2000 c; Einziger and 12 Beyer, 2007) indicates that a release (spalling off) of 15 percent of cladding CRUD may be 13 assumed as reasonably bounding to both normal and off

-normal conditions of storage, and a 14 release of 100 percent of the cladding CRUD is conservatively bounding to both postulated fire 15 and impact accidents during storage (NRC, 2014). 16 Fission Gases 17 The NRC's FRAPCON steady

-state fuel performance code has been previously used to assess 18 release fraction s of fission gases during transportation (NRC, 2011). The seven most common 19 fuel designs were evaluated using FRAPCON's modified Forsberg

-Massih model (8 x8, 9x9, 20 and 10x10 fuel for boiling water reactors (BWR s) and 14x14, 15x15, 16x16, and 17 x17 for 21 pressurized-water reactors (PWR s). For each fuel design, a number of different power histories 22 aimed at capturing possible realistic reactor irradiations were modeled. The fission gas content 23 within the free volume of the rods was evaluated for a total of 243 different cases (39 for each of 24 the BWR fuel designs

37 for 14x14 and 16x16 PWR fuel designs, and 26 for 15 x15 and 17x17 25 PWR fuel designs

). A review of the results indicates that a release of 15 percent of fission 26 3-9 gases may be assumed as reasonably bounding to normal conditions of transport scenarios for 1 rod average burnup s up to 62.5 GWd/MTU. The same release fraction may be reasonably 2 assumed for both normal and off

-normal conditions of storage.

3 During a fire accident scenario in storage, the fuel is not expected to reach temperatures high 4 enough that fission gases can diffuse out of the pellet matrix or grain boundaries to the rod 5 plenum. The thermal rupture tests showed that release occurred at higher temperatures than 6 those experience d during a transportation fire accident (NRC, 2000 c). The same behavior is 7 expected during a postulated fire accident condition of storage. Therefore, the same release 8 fraction of 15 percent of fission gases during normal/off

-normal conditions of storage may be 9 assumed to be reasonably bounding to the fire scenario under accident conditions of storage

. 10 In the case of postulated impact accident (drop) scenarios (e.g., during transfer or retrieval 11 operations), the pellet may be conservatively assumed to crumble. In this scenario, fission 12 gases retained within the pellet grain boundaries may be released in addition to those already 13 released from the fuel rod free volume (i.e., from the fuel-cladding gap and plenum). The 14 FRAPFGR model in FRAPCON may be used to predict the location of the fission gases within 15 the fuel pellet (NRC, 2011). The model has been validated with experimental data obtained 16 using an electron probe micro analyzer. The FRAPFGR model was used to calculate the 17 maximum fraction of the pellet-retained fission gases that may be released during a drop 18 impact, which was determined to be 20 percent. Therefore, assuming all fission gases within 19 the pellet gra in boundaries are released, a 35 percent (15 percent + 20 percent) maximum 20 release fraction may be assumed to be reasonably bounding to a postulated accident fire 21 scenario during storage. This value accounts for the 15 percent maximum fission gases 22 released from the fuel rod free volume (as calculated with the modified Forsberg

- Massih model) 23 and the 20 percent maximum fission gases released from the fuel pellet grain boundaries (as 24 calculated with the FRAPFGR model). These release fraction estimates are consistent with 25 previous NRC estimates (NRC, 200 0 c; NRC , 2007; Einziger and Beyer, 2007).

26 Volatiles 27 Mo st of the volatile release fractions originate from cesium

-based compounds in the form of 28 oxides or chlorides (NRC, 2000 c; NRC, 2014). These volatiles exhibit a different release 29 behavior in comparison to fission gases. Volatiles tend to migrate and aggregate at the rim on 30 the outer surface of the fuel pellet during reactor irradiation, which is characteristic of burnups 31 near or exceeding 60 GWd/MTU. The pellet rim is characterized by a fine crystalline grain 32 structure (0.1--0.3 µm or submicron in characteristic size) (Spino et al

., 2003; Einziger and 33 Beyer, 2007), a high porosity that may exceed 25 percent, and a high concentration of actinides 34 relative to the inner pellet matrix

. 35 Sandia National Laboratories assessed the maximum release fraction of volatiles (cesium and 36 other ruthenium-based compounds

) under drop and fire accident scenarios of transportation , 37 and determined it to be 0.003 percent (3x10-5) (NRC, 2000 c). This assessment included 38 modeling and analyses using various data from the literatur e. The volatile release fraction 39 during a fire accident scenario was determined to be lower than the release fraction during a 40 drop accident scenario (NRC, 2014; NRC, 2000 c). Therefore, a volatile release fraction of 41 0.003 percent (3 x 10-5) may be assumed to be reasonably bounding to normal, off

-normal, and 42 accident conditions of storage. This release fraction estimate is also consistent with an 43 independent estimate by Einziger and Beyer (2007).

44 45 3-10Fuel Fines 1 Release fractions from SNF fines during storage and transportation have been previously 2 documented (NRC, 2000 c; NRC, 2007; Benke et al., 2012; NRC, 2014). HBU SNF has a 3 different pellet microstructure than LBU SNF, which is characterized by an inner matrix and an 4 outer pellet rim layer. The thickness of the outer pellet rim layer increases with higher fuel 5 burnup. Therefore, differences in microstructure between the inner pellet matrix and the outer 6 pellet rim should be considered when evaluating release fractions of fuel fines from HBU SNF.

7 Although there is no reported literature on HBU SNF rim fracture as a function of impact energy , 8 other data can be used to indirectly assess the contribution of the rim layer to the release 9 fractions of fuel fines. Spino et al (1996) estimated the fracture toughness of the rim layer from 10 micro-indentation tests. Compared to the inner SNF matrix, the rim layer showed an increase of 11 fracture toughness. The increase of fracture toughness implies a decrease of release fraction. 12 Hirose et al (2015) also discussed results of axial dynamic impact tests simulating accident 13 conditions during transport, which are expected to be bounding to postulated drop scenarios 14 during dry storage. The dispersed particles from pellet breakage following impact were 15 collected and correlated to impact energy. The staff has compared the measured release 16 fraction of fuel fines from Hirose et al (2015) with previous NRC estimates of release fraction 17 versus impact energy for SNF and other brittle materials (depleted UO 2, glass and Synroc) (see 18 Figure 3 of NUREG 1864, "A Pilot Probabilistic Risk Assessment of a Dry Cask Storage System 19 at a Nuclear Power Plant" (NRC 2007)). Based on these analyses, the staff concludes that 20 there is no indication that pellet rim layer contributes to increased release fractions. 21 Since the outer HBU fuel pellet rim does not appear to contribute to additional release fractions, 22 previous NRC estimates for release fractions of fuel fines may continue to be used (NRC, 23 2000c; NRC, 2007; Benke, et al., 2012; Ahn et al., 2012; NRC, 2014). Per the range of 24 estimates in the literature, a release fraction for fuel fines of 0.003 percent (3x10-5) may be 25 assumed to be reasonably bounding to normal, off

-normal, and accident (drop impact) 26 conditions of storage. During a fire accident scenario, fuel oxidation is conservatively assumed 27 to increase the release fraction of fuel fines by a factor of 100 (NRC, 2000 c; Ahn et al 2012). 28 Therefore, a 0.3 percent (3x10-3) release fraction of fuel fines may be assumed as reasonably 29 bounding to fire accident conditions of storage.

30 The staff recognizes that various international cooperative research programs are currently 31 investigating release fractions from HBU SNF. Once those data are available to the public, the 32 staff will review and determine whether the conservative estimates in the above discussion 33 should be revis it ed. 34 3.2.3 Dry Storage U p To 20 Years 35 Section 1.2 discussed t he staff's review guidance for the licensing and certification of dry 36 storage of HBU SNF for a period of up to 20 year

s. The technical basis referenced in that 37 guidance supports the staff's conclusion that creep is not expected to result in gross rupture if 38 cladding temperatures are maintained below 400 °C (752 °F). 39 Chapter 2 also provi ded an asses m e of th e effects of hydri de reorientati on per static and 40 fatigue bending test results on HBU SNF specimens. Those test results provide a technical 41 basis for the staff's conclusion that the use of cladding mechanical properties (with either as- 42 irradiated or hydride- reoriented microstructure) is adequate for the structural evaluation of HBU 43 SNF when evaluating postulated drops during dry storage (e.g., drops during transfer 44 3-11 operations, non

-mechanistic DSS cask tipover). Refer to the current SRPs for dry storage of 1 SNF for staff review guidance on additional considerations for acceptable cladding

-only 2 mechanical properties (i.e., alloy type, burnup, temperature), on acceptable references for 3 cladding mechanical properties and on acceptance criteria for the structural evaluation of the 4 HBU fuel assembly for the drop accident scenarios. As indicated in Figure 3-1, supplemental 5 safety analyses are not expected for HBU SNF in dry storage for periods not exceeding 20 6 years. 7 3.2.4 Dry Storage Beyond 20 Years 8 As indicated in Figure 3-1, to address age

-related uncertainties related to the extended dry 9 storage of HBU SNF (i.e., dry storage beyond 20 years), the application is expected to be 10 supplemented with either results from a surrogate demonstration program or supplemental 11 safety analyses assuming justified hypothetical fuel reconfiguration scenarios. The results from 12 a surrogate demonstration program are meant to provide field

-obtained confirmation that the 13 fuel has remained in the analyzed configuration after 20 years of dry storage. If confirmation is 14 not provided, the safety analyses for the DSS should be supplemented to assume reconfigured 15 fuel. Consistent with the requirements in 10 CFR Part 72, the supplemental information may be 16 provided in either the initial license or CoC application (per 10 CFR 72.4 0(a) and 17 10 CFR 72.238, "Issuance of an NRC Certificate of Compliance"

) or in a renewal application 18 (10 CF R 72.42(a) and 10 CFR 72.240(a))

. 19 The NRC has approved the licensing and certification of HBU SNF for an initial 20

-year-term p er 20 the technical basis in the staff's review guidance, as discussed in Section 1.2. However, t h e 21 staff has recognized that the technical basis is based on short

-term accelerated creep testing 22 (i.e., laboratory scale testing up to a few months), which results in increased uncertainties when 23 extrapolated to long periods of dry storage

- see Appendix D to NUREG-1927, Revision 1 24 (NRC, 2016 b). Although the staff has confidence based on this short

-term testing that creep

-25 related degradation of the HBU fuel will not adversely affect its analyzed configuration for 26 storage periods beyond 20 years, there is no operational field-obtained data to confirm this 27 expectation, as was done in the prior demonstration on LBU fuel described in NUREG/CR

-6745, 28 "Dry Cask Storage Characterization Project

-Phase 1; CASTOR V/21 Cask Opening and 29 Examination," issued September 2001 (NRC, 2001),; and NUREG/CR 6831, "Examination of 30 Spent PWR Fuel Rods after 15 Years in Dry Storage," issued September 2003 (NRC, 2003b).

31 In addition, the staff also acknowledges that while the CIRFT results obtained to

-date (as 32 discussed in Chapter 2) provide an adequate technical basis for assessing the separate effects 33 of hydride reorientation, the results do not account for potential synergistic effects of various 34 physical and chemical phenomena occurring during extended dry storage (e.g., cladding creep, 35 hydride reorientation, irradiation hardening, oxidation, hydriding caused by residual water 36 hydrolysis, etc.

- see NUREG-2214, "Managing Aging Processes in Storage (MAPS) Report,"

37 issued October 2017 (NRC, 2017c) for discussions on these phenomena

). Therefore, the staff 38 considers it prudent to gather and review evidence that HBU fuel in dry storage beyond 20 39 years has maintain ed its analyzed configuration be gathered and reviewed.

40 3.2.4.1 Supplemental Results from Confirmatory Demonstration 41 A demonstration program, like that conducted for LBU SNF (NRC, 2003; NRC , 2001; NRC, 42 2003 b), may be used to confirm the results from separate

-effects testing, which has provided 43 the technical bases for dry storage of HBU SNF beyond 20 years

. 44 3-12 3.2.4.1.1 Initial Licensing or Certification 1 Consistent with 10 CFR 72.42(a) and 10 CFR 72.238, an applicant may request approval for dry 2 storage of HBU SNF for periods up to 40 years. These applications are not required to provide 3 aging management programs (AMPs), as these programs are expected only in renewal 4 applications. Instead, for initial license s and CoC approvals for dry storage beyond 20 years (up 5 to 40 years), the application may describe the activities to obtain and evaluate confirmatory data 6 from a demonstration program under the aegis of a maintenance plan. The maintenance plan 7 would be implemented after the initial 20 years of dry storage. Applicants may refer to 8 Appendices B and D to NUREG-1927, Revision 1 (NRC, 2016 b) when developing the 9 description of activities to assess data from the confirmatory demonstration

. 10 3.2.4.1.2 Renewal Applications 11 Consistent with 10 CFR 72.42(a) and 10 CFR 72.240(a), a renewal application for a specific 12 license or CoC, may describe the activities to obtain and evaluate confirmatory data to be 13 performed under the aegis of an AMP.

Applicants may refer to Appendices B and D to NUREG-14 1927, Revision 1 (NRC, 2016 b) when developing the description of activities to assess data 15 from the confirmatory demonstration

. 16 3.2.4.2 Supplemental Safety Analyses 17 As an alternative approach to a confirmatory demonstration for HBU SNF, an application may 18 supplement the design bases with safety analyses that demonstrate the DSS can still meet th e 19 pertinent regulatory requirements by assuming hypothetical reconfiguration of the HBU fuel 20 contents into justified geometric form

s. This alternative approach would demonstrate that the 21 design-bases fuel, even if reconfigured

, can still meet the 10 CFR Part 72 requirements for 22 thermal, confinement, criticality safety and shielding during normal, off

-normal, and accident 23 conditions.

For renewal applications, a separate license amendment or CoC amendment may 24 be required if the changes in the supplemental safety analyses do not meet the acceptance 25 criteria in 10 CFR 72.48, "Changes, Tests, and Experiments."

. 26 In NUREG/CR-7203 (NRC, 2015), ORNL Oak Ridge National Laboratory (ORNL) evaluated the 27 impact of a wide range of postulated fuel reconfiguration scenarios under non

-mechanistic 28 causes of fuel assembly geometry change with respect to criticality, shielding (dose rates), 29 containment, and thermal. The study considered three fuel reconfiguration categories , which 30 were characterized by either category 1, cladding failure

category 2, rod/assembly deformation 31 without cladding failure

or category 3 changes to assembly axial alignment without cladding 32 failure. Within configurations in both Categor y 1 and Category 2, the study identified various 33 scenarios

34 Category 1: cladding failure 35 - Scenario 1(a): breached rods 36 - Scenario 1(b): damaged rods 37 38 39 3-13 Category 2:

rod/assembly deformation without cladding failure 1 - Scenario 2(a): configurations associated with side drop 2 - Scenario 2(b): configurations associated with end drop 3 Category 3: changes to assembly axial alignment without cladding failure 4 The analyses in NUREG/CR

-7203 (NRC, 2015) considered representative SNF transportation 5 packages, and a range of fuel initial enrichments, discharge burnup values, and decay times.

6 Two package designs were analyzed: a general burnup credit (GBC)

-32 package containing 32 7 PWR fuel assemblies and a GBC

-68 package containing 68 BWR fuel assemblies. Although 8 NUREG/CR-7203 did not evaluate reconfiguration in DSSs, the scenarios and analytical 9 methods may also be applicable to those designs, as the loads experienced during transport 10 conditions (normal, hypothetical accident) are expected to bound those experienced during 11 storage (normal, off

-normal and accident). The results in NUREG/CR

-7203 should not be 12 assumed to be generically applicable as fuel reconfiguration may have different consequences 13 for a DSS design other than the generic models evaluated in the study. However, the following 14 sections discuss considerations in developing supplemental safety analyses for other DSS 15 designs according to the reconfiguration scenarios considered in NUREG/CR

-7203. 16 3.2.4.2.1 Materials and Structural 17 An application relying on supplemental safety analyses based on hypothetical reconfiguration of 18 the HBU SNF contents should provide a structural evaluation for the package and its fuel 19 contents using any of the approaches discussed in Section 3.2. The staff will review the 20 structural evaluation and the assumed material mechanical properties, including any changes 21 due to higher temperatures resulting from fuel reconfiguration, in a manner consistent with the 22 guidance in the current SRP for dry storage of SNF

. 23 3.2.4.2.2 Confinement 24 An application relying on supplemental safety analyses based on hypothetical reconfiguration of 25 the HBU SNF is expected to demonstrate that the DSS design meets the regulatory 26 requirements for confinement if data from a surrogate demonstration program, used for 27 confirmatory demonstration following the guidance in NUREG

-1927, Revision 1 (NRC, 2016 b), 28 is not available before the renewal of the license for previously dry

-stored fuel for periods longer 29 than 20 years

. 30 However, if the thermal, structural, and material analyses , together with aging management 31 activities for the DSS subcomponents supporting confinement (i.e., confinement boundary) 3 , are 32 used to provide assurance that the allowable leak rate is maintained even after hypothetical 33 reconfiguration of the fuel under normal, off-normal and accident

-level conditions, supplemental 34 safety analysis for the confinement performance of the DSS design are not expected. Thermal 35 analyses demonstrate that all DSS subcomponents supporting confinement (i.e., confinement 36 3 Aging management activities may be conducted under either the aegis of an NRC

-approved AMP (for renewal applications) or a maintenance plan (for initial license or CoC applications requesting approval for periods exceeding 20 years).

3-14 boundary) will be able to withstand their maximum operating temperatures and pressures under 1 normal, off

-normal and accident-level conditions

. 2 3.2.4.2.3 Thermal 3 Fuel reconfiguration can affect the efficiency of heat removal from the fuel because of changes 4 in (1) thermo-physical properties of the canister gas space stemming from release of fuel rod 5 inert gas and fission product gases, (2) heat source location within the canister, and (3) changes 6 in flow area (convection), conduction lengths (conduction) and radiation view factors (thermal 7 radiation). As part of a defense-in-depth approach for addressing a ge-related uncertainties for 8 uncanned and undamaged HBU fuel in dry storage beyond 20 years, the thermal analyses 9 would be expected to analyze scenarios for normal, off

-normal , and accident conditions of 10 storage by assuming the fuel may become substantially altered. NUREG/CR-7203 (NRC, 2015) 11 describes the impact on the DSS canister pressure and the fuel cladding and DSS component 12 temperatures for various scenarios of fuel geometry changes. These are examined below. In 13 general, the results in NUREG/CR

-720 3 should not be considered generically applicable. The 14 thermal analyses of the application are expected to consider scenarios discussed in 15 NUREG/CR-7203 to determine consistency in the analytical methods, scenario phenomena, 16 and results

. The thermal analyses are expected to assess the impact of the fuel reconfiguration 17 on the fuel cladding and DSS component temperatures and the canister pressure for the 18 particular DSS design. 19 For Scenario 1(a) in Category 1 (see Section 3.2.4.2)

, the fuel rods are assumed to breach in 20 such a manner that the cladding remains in its nominal geometry (no fuel reconfiguration), but 21 depending on the canister orientation (horizontal or vertical), the release of fuel rod fill gas and 22 fission product gases can cause a change to component peak temperatures. For Scenario 1(b) 23 in Category 1 , for configurations where a n assembly (or assemblies) is represented as a debris 24 pile(s) inside its basket cell, fuel reconfiguration has a larger impact on the component 25 temperatures for the vertical orientation than for the horizontal orientation, but the packing 26 fraction of the debris bed has minor impact on the component temperatures.

For both 27 Scenarios 1(a) and 1(b), release of the fuel rod gaseous contents increases the number of 28 moles of gas and therefore increases the canister pressure. The canister pressure is expected 29 to increase with the increased fuel rod release fractions. 30 For Scenarios 2(a) and 2(b), the fuel rods are assumed to remain intact without gaseous 31 leakage into the canister space. T he changes of the fuel assembly lattice (contraction in 32 Scenario 2(a) and expansion in Scenario 2(b)) could cause either an increase or decrease in 33 the component temperatures of the storage system depending on the initial assembly geometry 34 and whether the storage system relies on convection for heat transfer. In general, scenarios 35 Scenario 2(a) and Scenario 2(b) have minor impact on the fuel cladding and DSS component 36 temperatures and canister pressure. For Category 3 , the fuel rods are assumed to remain intact 37 without gaseous leakage into the canister space, but the axial shifting of the assembly changes 38 the heat source location within the canister. Changes in assembly axial alignment within the 39 basket cells are expected to have minor impact on the component temperatures and the 40 canister pressure.

41 Normal, Off-Normal, and Accident Conditions of Storage 42 Based on the thermal phenomena described above and NUREG/CR

-7203 (NRC, 2015), an 43 approach acceptable to staff would evaluate the impact of Scenarios 1(a) and 1(b) on the 44 canister pressure and the fuel cladding and package component temperatures assuming 45 3-15 rupture of 1 percent, 10 percent and 100 percent of the fuel rods for normal, off

-normal, and 1 accident conditions, respectively.

2 Although Scenarios 2(a) and 2(b) in Category 2 and Category 3 are not expected to have a 3 significant impact on DSS thermal performance under normal, off

-normal and accident 4 conditions, because the fuel rods in Scenarios 2(a), 2(b) and 3 are assumed to remain intact 5 without gaseous leakage into the canister space

, the applicant may need to provide a thermal 6 evaluation due to specifics of the DSS design

. 7 3.2.4.2.4 Criticality 8 An application may demonstrate that a DSS meets the regulatory requirements for criticality 9 safety for the period beyond 20 years by assuming hypothetical reconfiguration of the HBU 10 SNF into a bounding geometric form. This approach is one way to ensure compliance with 11 10 CFR 72.124, "Criteria for Nuclear Criticality Safety,"

or 10 CFR 72.236(c) during normal, off-12 normal, and accident conditions

, if the structural evaluation does not adequately define the 13 mechanical properties of the cladding

. 14 As mentioned previously, ORNL examined hypothetical fuel reconfiguration for various 15 scenarios and the impacts on the criticality safety of a DSS and documented the results in 16 NUREG/CR-7203. This study, considers burnup up to 70 GWd/MTU for criticality evaluations.

17 NUREG/CR-7203 provides some insight into the reactivity trends for various reconfiguration 18 scenarios; howeve r the results in NUREG/CR

-7203 (NRC, 2015) should not be considered 19 generically applicable with respect to criticality safety analyses. 20 Criticality is not a concern for dry SNF systems, as SNF requires moderation to reach criticality. 21 Although DSS casks are expected to remain dry while in storage, cask users may be allowed to 22 load and unload a cask in a wet environment. The criticality analyses in NUREG/CR

-7203 are 23 performed with an assumption of fully flooded conditions and any conclusions adopted are 24 applicable to analyses that support wet loading and unloading. The following considerations for 25 criticality evaluations for reconfigured fuel are applicable only to DSS scenarios where there 26 may be flooding within the canister. Otherwise, the staff does not find reconfiguration to pose a 27 criticality safety concern for a dry system.

28 All of the criticality safety analyses presented in NUREG/CR

-7203 take credit for burned fuel 29 nuclides (burnup credit) and the conclusions may not be applicable to criticality analyses that 30 assume a fresh fuel composition. In its review of the burnup credit methodology and code 31 benchmarking used to support a criticality safety evaluation, the staff will follow the guidance in 32 ISG-8, Revision 3, "Burnup Credit in the Criticality Safety Analyses of PWR Spent Fuel in 33 Transportation and Storage Casks," issued in September 2012 (NRC, 2012) to review the 34 burnup credit analys e s. ISG-8, Revision 3, does not endorse any particular methodology for 35 BWR fuel burnup credit. The staff does not necessarily endorse the methodology described in 36 NUREG/CR-7203 for BWR fuel DSS, and considers it to be for illustration only.

37 For criticality safety analyses using burnup credit, NUREG/CR

-7203 (NRC, 2015) shows that 38 reactivity increases for longer decay times (e.g., analyses supporting storage beyond 20 years

) 39 would need to use an appropriate decay time within the criticality evaluations.

The enrichment 40 and burnup values assumed within the criticality evaluations in NUREG/CR

-7203 may differ 41 from those allowed within another storage system. However NUREG/CR

-7203 states that no 42 significant differences were observed in trends between configurations that evaluated fuel at 43 44.25 GWd/MTU and 70 GWd/MTU.

44 3-16 The following sections discuss information from NUREG/CR-7203 that may be applicable when 1 performing reconfiguration analyses within a criticality evaluation for HBU fuel under normal, off

-2 normal , and accident conditions of storage.

3 Normal Conditions of Storage 4 In an approach acceptable to the staff, t he applicant's criticality safety analys e s would consider 5 the reactivity impact of 3-percent fuel failure during normal conditions of storage. The most 6 applicable scenario from NUREG/CR

-7203 (NRC, 2015) is Scenario 1(a)

(See Section 3.2.4.2 7 above for a description of the scenarios

). 8 ORNL created Scenario 1(a) to represent breached rods. ORNL assumed that a percentage of 9 the rods were breached and that cladding from these rods failed completely and then removed 10 this percentage of fuel rods from the system.

This is conservative as SNF systems are 11 undermoderated and replacing fuel with moderator typically causes reactivity to increase. Using 12 a fresh fuel composition for PWR fuel, ORNL's models in NUREG/CR

-7203 showed that 13 reactivity decreases when removing r ods. Therefore, this type of analysis may not be 14 appropriate for PWR analyses that assume a fresh fuel composition. The location assumed for 15 failed or removed rods can significant ly effect reactivity. ORNL showed in Section A.1.1 of 16 NUREG/CR-7203 that removing rods from the center of the assembly causes reactivity to 17 increase the most.

18 In NUREG/CR-7203, ORNL also showed the number of rods removed that produces the 19 maximum reactivity. For the systems studied , NUREG/CR-7203 shows that the maximum 20 reactivity occurs when a number of rods far greater than 1

-percent is removed from the system.

21 NUREG/CR-7203 also presents the results of a sensitivity study showing that reactivity increases 22 even more for Scenario 1(a) when it is assumed that the failed fuel relocates to a location outside 23 of the absorber plate. This is based on the generic system s modeled for the study. A different 24 system may allow relocation of the failed rod material outside of the absorber plate material to a 25 different extent.

26 Off-Normal Conditions of Storage 27 In an approach acceptable to the staff, t he applicant's criticality safety analys e s would consider 28 the reactivity impact of 10-percent fuel failure under off

-normal conditions of storage. The 29 methods discussed in the previous section on normal conditions of storage also apply to off

-30 normal conditions of storage

however the applicant would consider fuel failure up to 10 percent 31 rather than 1 percent. Scenario 1(a) can be used to represent rod failure via removing rods 32 from the system.

In this case an applicant would remove 10-percent of the rods rather than 1-33 percent. The applicant would remove rods in such a way that it produces maximum reactivity 34 and consider relocation of the fuel to outside of the absorber plates.

35 Accident Conditions of Storage 36 In an approach acceptable to the staff, t he applicant's criticality safety analys e s would consider 37 the reactivity impact of 100-percent fuel failure under accident conditions of storage. The 38 damaged fuel models in Section A.1.2 for Scenario 1(b) from NUREG/CR

-7203 are applicable 39 when representing 100 percent failed fuel.

40 3-17 Scenario 1(b) from NUREG/CR

-7203 considers reconfiguration of damaged fuel. With 100-1 percent compromise in cladding integrity, reconfiguration is considered to the maximum extent.

2 Section A.1.2 of NUREG/CR

-7203 shows that a model assuming an "ordered pellet array" is 3 more reactive than a homogenous mixture of fuel, cladding materials and water.

4 3.2.4.2.5 Shielding 5 An application may demonstrate that a DS S continues to meet the regulatory dose limits for the 6 period beyond 20 years by assuming hypothetical reconfiguration of the HBU SNF into a 7 justified bounding geometric form under normal, off

-normal, and accident conditions. This 8 method is one way to demonstrate compliance with 10 CFR 72.104, 10 CFR 72.106, or 10 CFR 9 72.236(d). 10 To assess the impacts of various fuel geometry changes on the shielding designs of DSSs and 11 ISFSI s, ORNL analyzed various scenarios of fuel geometry changes and the impact on the 12 annual dose at the ISFSI boundary and dose rates near the cask and presented the results in 13 NUREG/CR-7203 (NRC, 2015)

. 14 Appendix B to NUREG/CR-7203 (provides some insight into the effects on external dose for 15 various reconfiguration scenarios

however the results in NUREG/CR

-7203 should not be 16 considered generically applicable with respect to external dose and dose rate evaluations. A 17 DSS designer would assess the impacts of fuel reconfiguration on external dose and dose rates 18 for its particular design. 19 This section discusses an approach acceptable to the staff for addressing the impacts on 20 external dose and dose rates when considering possible reconfiguration of HBU fuel for a period 21 of storage beyond 20 years. This discusses the scenarios from NUREG/CR

-7203 most 22 applicable to the reconfiguration under normal, off

-normal , and accident conditions of storage as 23 well as the analytical assumptions likely to result in bounding dose and dose rates based on the 24 results from NUREG/CR

-7203. The NUREG has considered burnup up to 65 GWd/MTU within 25 its dose and dose rate evaluations. As discussed in Section B.5 of NUREG/CR

-7203, different 26 nuclides become important to external dose and dose rate based on the decay time.

27 Since reconfiguration is to be considered after 20 years of storage, and this length of cooling 28 time is generally much longer than cooling times used to establish loading tables, applicants 29 may be able to make the justification that increases to external dose due to reconfiguration are 30 bounded by the additional cooling time the assemblies will experience.

31 NUREG/CR-7203 also indicates that fuel assembly type, (i.e., PWR vs BWR

), may have a 32 significant impact on the surface dose rate and controlled area boundary dose under fuel 33 reconfiguration scenarios. Table s 13 and 14 of NUREG/CR

-7203 show the difference in dose 34 rate increase for BWR and PWR SNF. A DSS system may permit storage of other fuel 35 assemblies, with different allowable burnup and enrichments to which the results of 36 NUREG/CR-7203 (NRC, 2015) do not apply. The burnup profile and depletion parameters used 37 to create the source term within NUREG/CR

-7203 may also not be generically applicable.

38 Normal Conditions of Storage 39 In an approach acceptable to the staff, t he applicant's external dose and dose rate evaluation 40 would consider the impact of 3-percent fuel failure during normal conditions of storage. The 41 most applicable scenario from NUREG/CR

-7203 is Category 1, fuel failure, Scenario, 1(a). If 42 3-18 cladding is breached and the fuel fails, this could lead to source relocation or change of the 1 geometric shape of the source. Based on NUREG/CR

-7203, the impact on the controlled

-area 2 boundary dose caused by source relocation resulting from 1-percent fuel failure is insignificant.

3 For a different DSS, the application may need to discuss potential fuel failure and source 4 reconfiguration and the potential impact on controlled

-area boundary doses as required by 10 5 CFR 72.104 and 10 CFR 72.106. 6 Depending on the DSS and the resultant fuel geometry, the dose rate may increase significantly 7 as the detector moves close to the cask. Although it may not cause a significant change to the 8 dose and therefore may not constitute a significant concern for people at the controlled area 9 boundary, the changes of source term geometry wi ll affect the doses of occupational workers 10 who need to perform necessary work around the casks.

In general, an application should 11 consider the impact of HBF failure on the near cask dose rate and potential impacts on radiation 12 protection associated with ISFSI surveillance and maintenance operations.

13 Off-Normal Conditions of Storage 14 In an approach acceptable to the staff, t he applicant's external dose and dose rate evaluation 15 for HBF would consider the impact of 10-percent fuel failure under off-normal conditions of 16 storage. If cladding is breached and fails, the fuel, and hence the source, may relocate to 17 different parts of the fuel basket. The impact of HBFfailure on dose at the controlled

-area 18 boundary for storage under off

-normal conditions of dry storage operations should be examined.

19 A 10-percent fuel failure is similar to Scenario 1(a) in NUREG/CR

-7203 (NRC, 2015). For 20 Scenario 1(a), breached rods, ORNL assumed the rods turned to rubble and calculated the 21 dose rate when the fuel mixture relocated to the bottom of the fuel assembly. ORNL assumed 22 failure of 10-percent of fuel rods collected into the available free volume within the assembly 23 lower hardware region. Section B.4.1 of NUREG/CR

-7203 discusses the implementation in 24 detail. ORNL reduced the source strength and density of the active fuel zone by the failure 25 percentage and relocated this source to the bottom of the fuel assembly and increased the 26 source strength and density accordingly. The storage system in NUREG/CR

-7203 is modeled 27 as a vertically-oriented storage system. Fuel would likely not relocate this way in a horizontal 28 storage system

, and the models is not necessarily applicable to a horizontal system.

29 In Section B.5.5 of NUREG/CR

-7203 , ORNL discuss the results of the study performed on the 30 individual storage, which shows that there could be significant increases in the dose rate near 31 the cask. It concludes that fuel configuration changes can cause significant dose rate increases 32 relative to the nominal intact fuel configuration in the cask outer regions that face air vent 33 locations. NUREG/CR-7203 states that the change in radiation dose rate a way from air vent 34 locations is either small or negligible.

35 Similar to normal conditions of storage, the changes in source term geometry will impact the 36 doses of occupational workers who need to perform necessary surveillance and maintenance 37 work around the casks. To assess the impacts on radiation protection, an applicant may need 38 to evaluate the surface dose rate increase resulting from reconfiguration.

39 Accident Conditions of Storage 40 In an approach acceptable to the staff, t he applicant's external dose and dose rate evaluation 41 for HBF would consider the impact of 100-percent fuel failure during normal conditions of 42 storage. If cladding is breached and the fuel fails, this may cause the fuel, and hence the 43 3-19 source, to relocate to different parts of the fuel basket. Based NUREG/CR

-7203 (NRC, 2015), 1 the impacts on the controlled

-area boundary dose caused by source relocation resulting from 2 100 percent fuel failure w ill result in significant increases in the dose rate near the cask and 3 annual dose on the controlled area boundary

. Scenarios 1(b) and 2 in NUREG/CR 7203 can 4 represent 100-percent fuel failure. 5 At the controlled area boundary, 100-percent fuel reconfiguration can have a significant impact 6 on the annual dose. It can also significantly affect the dose rate near the cask and the radiation 7 protection associated with ISFSI remediation operations. Table s B.9 and B.10 of Appendix B t o 8 NUREG/CR-7203 (NRC, 2015) show the relative changes in dose rates at 1 meter from a 9 sample PWR fuel cask and a sample BWR fuel cask, respectively. Table B.11 of Appendix B to 10 NUREG/CR-7203 shows the estimated relative impact on controlled

-area boundary dose from 11 fuel reconfiguration. The data presented in these tables show that the impacts on the dose 12 rates at the cask side, particularly the dose rate near the vent ports are significant.

13 In Scenario 1(b), ORNL assumed that the assembly and basket plate material is homogenized , 14 placed it at the bottom of the cask

, and determined that the limiting packing fraction is 0.58.

15 This scenario did not produce an increase in site boundary dose

however, it did show an 16 increase in local dose rates. The location of the "bottom" of the cask would depend on whether 17 the DSS is vertical or horizontal. Homogenizing the basket material with the fuel rubble may be 18 overly conservative for a horizontal configuration, and applicants may choose to maintain basket 19 integrity similar to the Scenario S2 model in Section B.4.2 of NUREG/CR-7203 when evaluating 20 dose or dose rates for a horizontal system or a tip-over scenario.

21 For Scenario 1(b), ORNL also assumed that the fuel and basket material forms a homogenized 22 rubble that is distributed throughout the canister cavity. This scenario produced an increase in 23 site boundary dose.

24 3.3 Canned Fuel (Damaged Fuel) 25 10 CFR 72.122(h)(1) requires SNF, including HBU , with gross ruptures (i.e., classified as 26 damaged) be placed in a can designed for damaged fuel or in an acceptable alternative.

The 27 staff will follow the guidance in the current SRPs for dry storage of SNF in its review of an 28 application for a DSS with damaged HBU SNF contents

. 29 NUREG-2224 Dry Storage and Transportation of High Burnup Spent Nuclear Fuel Draft Report for Comment Offic e of N u clear Material Safety a n d Safeg u ards NRC Reference Material As of November 1999, you may electronically access NUREG-series publications and other NRC records at the http://www.nrc.gov/reading

-rm.html. Publicly released records include, to name a few, NUREG

-series publications; Federal Register notices; applicant, licensee, and vendor documents and correspondence; NRC correspondence and internal memoranda; bulletins and information notices; inspection and investigative reports; licensee event reports; and Commission papers and their attachments.

NRC publications in the NUREG series, NRC regulations, and Title 10, Energy ,Code of Federal Regulations may also be purchased from one of these two sources. 1. The Superintendent of DocumentsU.S. Government Publishing OfficeWashington, DC 20402-0001Internet: http://bookstore.gpo.govTelephone:

1-866-512-1800Fax: (202) 512-2104 2. The National Technical Information Service5301 Shawnee RoadAlexandria, VA 22161-0002http://www.ntis.gov 1-800-553-6847 or, locally, (703) 605-6000A single copy of each NRC draft report for comment is available free, to the extent of supply, upon written request as follows:

U.S. Nuclear Regulatory Commission Office of Administration Multimedia, Graphics and Storage & Distribution Branch Washington, DC 20555-0001 E-mail: distribution.resource@nrc.govFacsimile:

(301) 415-2289Some publications in the NUREG series that are posted at the http://www.nrc.gov/reading

-rm/doc-collections/nuregs are updated periodically and may differ from the last printed version. Although references to material found on a Web site bear the date the material was accessed, the material available on the date cited may subsequently be removed from the site.

N on-NRC Reference Material Documents available from public and special technical libraries include all open literature items, such as books, journal articles, transactions, Federal Register notices, Federal and State legislation, and congressional reports. Such documents as theses, dissertations, foreign reports and translations, and non

-NRC conference proceedings may be purchased from their sponsoring organization.

Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at The NRC Technical Library Two White Flint North 11545 Rockville Pike Rockville, MD 20852-2738 These standards are available in the library for reference use by the public.

Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from American National Standards Institute 11 West 42 nd Street New York, NY 10036

-8002 http://www.ansi.org (212)642-4900 AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATION S Legally binding regulatory requirements are stated only in laws; NRC regulations; licenses, including technical specifications; or orders, not in NUREG

-series publications. The views expressed in contractor

-prepared publications in this series are not necessarily those of the NRC.

The NUREG series comprises (1) technical and administrative reports and books prepared by the staff (NUREG-XXXX) or agency contractors (NUREG/CR-XXXX), (2) proceedings of conferences (NUREG/CP-XXXX), (3) reports resulting from international agreements (NUREG/IA-XXXX), (4) brochures (NUREG/BR

-XXXX), and (5) compilations of legal decisions and orders of the Commission and (NUREG-0750). DISCLAIMER:

This report was prepared as an account of work sponsored by an agency of the U.S. Government. Neither the U.S. Government nor any agency thereof, nor any employee, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any information, apparatus, product, or process disclosed in this publication, or represents that its use by such third party would not infringe privately owned rights.

SR-CR 10/2017 Dry Storage and Transportati on of High Burnup

Spent N uclear Fuel Draft R eport f or Comment M a nu script Completed:

April 2018 Dat e P u blis h ed: J u ly 2018 Prepared b y: T.A hn H.A kh a v a nn i k G.Bjorkman F.C. C h a ng W.Reed A.Rigato D.T a ng R.D. Torres B.H. White

V.W i lson O f f i c e o f N u c l e ar Mat e rialSa f e ty and Sa f eg uar d s NUREG-2224 C O M M E N T S O N D RA F T R E P O RT Any interest ed party may submit comments on this report f or considerati on by t he staff of t he U.S. Nuclear Regulatory Commission (NRC).

Comments m ay be accompanied by additional relevant informati on or supporti ng data. Please specify t he report num ber NUREG-222 4 in your comments, and send them by the end of the comment peri od specified in t he Federal Register notic e announci ng t he availability of this report. Addr e ss e s: Y ou may submit comments by any one of t he followi ng methods. Please include Docket I D NRC-2 018-0 066 i n the subject l i ne of y our comments. Comments submitt ed in writing or in electronic form will be posted on the NRC W eb site and on t he Federal rulemaking Web si t e (http://www.regulations.gov

). Federa l R ule mak ing Web site: G o t o http://www.regulations.gov and searc h for documents fil ed under D ocket I D NRC-2 018-0 066. Address ques tions about NRC doc kets to J ennifer Borges; t elephone: 301-2 87-9 127; e-m ail: Jennifer.Borges@nrc.govFor any questions about the material i n this r eport, pl ease contac t: W endy R eed, 301-4 15-7 213 or, e-m ail at W endy.Reed@nrc.gov. Please be aware that any comments that y ou submit to t he NRC w ill be consider ed a public record and entere d int o the Agencywi de Documents Access and M anagement System. D o not provi de informati on y ou woul d not want t o be publicly available.

ABSTRACT 1 Time-dependent changes on the cladding performance of high burnup (HBU) spent nuclear fuel 2 (SNF) are all primarily driven by the fuel's temperature, rod internal pressure (and 3 corresponding pressure

-induced cladding hoop stresses), and the environment during dry 4 storage or transport operations. Historically, the potential for these changes to compromise the 5 analyzed fuel configuration in dry storage systems and transportation packages has been 6 addressed through safety review guidance. This guidance defines adequate fuel conditions, 7 including peak cladding temperatures during short

-term loading operations to prevent or 8 mitigate degradation of the cladding. The purpose of this report is to expand the technical basis 9 in support of that guidance, as it pertains to the mechanism of hydride reorientation in HBU SNF 10 cladding. 11 Hydride reorientation is a process in which the orientation of hydrides precipitated in HBU SNF 12 cladding during reactor operation change s from the circumferential

-axial to the radial

-axial 13 direction. Research results over the last decade have shown that hydride reorientation can still 14 occur at temperatures and stresses lower than those assumed in the current staff review 15 guidance. Therefore, the U.S. Nuclear Regulatory Commission (NRC) has since sponsored 16 additional research to better understand whether hydride reorientation could affect the 17 mechanical behavior of HBU SNF cladding and compromise the fuel configuration analyzed in 18 dry storage systems and transportation packages.

19 This report provides an engineering assessment of the results of research on the mechanical 20 performance of HBU SNF following hydride reorientation. Based on the conclusions of that 21 assessment, the report then presents example approaches for licensing and certification of HBU 22 SNF for dry storage (under Title 10 of the Code of Federal Regulations (10 CFR) Part 72 , 23 "Licensing Requirements for the Independent Storage of Spent Nuclear Fuel and High

-Level 24 Radioactive Waste, and Reactor

-Related Greater Than Class C Waste") and transportation 25 (under 10 CFR Part 71 , "Packaging and Transportation of Radioactive Material"

). 26 The information in this report is not intended for use in applications for wet storage facilities or 27 monitored retrievable storage installations licensed under 10 CFR Part 72.

28 Nothing contained in this report is to be construed as having the force or effect of regulations.

29 Comments regarding errors or omissions, as well as suggestions for improvement of this 30 NUREG should be sent to the Director, Division of Spent Fuel Management, U.S. Nuclear 31 Regulatory Commission, Washington, D.C., 20555

-0001. 32 Paperwork Reduction Act 33 34 This NUREG provides guidance for implementing the mandatory information collections in 10 35 CFR Parts 71 and 72 that are subject to the Paperwork Reduction Act of 1995 (44 U.S.C. 3501 36 et. seq.). These information collections were approved by the Office of Management and Budget 37 (OMB) under control numbers 3150

-0008 and 3150

-0132. Send comments regarding this 38 information collection to the Information Services Branch, U.S. Nuclear Regulatory Commission, 39 Washington, DC 20555

-0001, or by e

-mail to Infocollects.Resource@nrc.gov, and to the Desk 40 Officer, Office of Information and Regulatory Affairs, NEOB

-10202, (3150

-0008, 3150

-0132) 41 Office of Management and Budget, Washington, DC 20503.

42 43 Public Protection Notification 1 The NRC may not conduct or sponsor, and a person is not required to respond to, a collection 2 of information unless the document requesting or requiring the collection displays a currently 3 valid OMB control number.

4 v CONTENTS 1 ABSTRACT ...................................................................................................................

iii 2 CONTENTS .................................................................................................................... v 3 LIST OF FIGURES

........................................................................................................

vii 4 LIST OF TABL ES. .........................................................................................................

ix 5 ACKNOWLEDGMENTS

................................................................................................

xi 6 ABBREVIATIONS AND ACRONYMS .........................................................................

xiii 7 1 INTRODUCTION

...................................... 8 1.1 Back ground ...........................................................

9 1.2 Fuel Cladding Performance and Staff's Review Guidance

................................1-2 10 1.3 Cladding Creep .................................................................................................1-4 11 1.4 Effects of Hydrogen on Cladding Mechanical Performance

...............................1-5 12 1.5 Hydride Reorientation

.......................................................................................1-7 13 1.5.1 Hydride Dissolution and Precipitation

....................................................1-8 14 1.5.2 Fuel Cladding Fabrication Process

...................................................... 1-10 15 1.5.3 End-Of-Life Rod Internal Pressures and Cladding Hoop 16 Stresses .............................................................................................. 1-11 17 1.5.3.1 End-Of-Life Rod Internal Pressures for Pressurized

-18 Water Reactor Fuel Rods

..................................................... 1-11 19 1.5.3.2 Gas Temperatures for Fuel Rods During Drying

-20 Transfer, Storage and Transportation

................................... 1-14 21 1.5.3.3 Peak Cladding Hoop Stresses for Pressurized

-Water 22 Reactor Fuel Rods During Drying

-Transfer and 23 Storage/Transport Operations

.............................................. 1-15 24 1.5.4 Ring Compression Testing

.................................................................. 1-18 25 1.5.5 Staff's Assessment of Ring Compression Testing Results

................... 1-24 26 2 ASSESSMENT OF STATIC BENDING AND FATIGUE STRENGTH 27 RESULTS ON HIGH BURNUP SPENT NUCLEAR FUE L................................ 2-1 28 2.1 Introduction

.......................................................................................................2-1 29 2.2 Cyclic Integrated Reversible Fatigue Tester

......................................................2-1 30 2.3 Application Of The Static Test Results

..............................................................2-4 31 2.3.1 Spent Fuel Rod Behavior in Bending. ....................................................2-5 32 2.3.2 Composite Behavior o f a Spent Fuel Rod. .............................................2-6 33 2.3.3 Calculation of Cladding Strain Using Factored Cladding-Only 34 Properties ............................................................................................ 2-10 35 2.3.3.1 Two Alternatives for Calculating Cladding Stress a nd 36 Strain During Drop Accidents

................................................ 2-13 37 2.3.4 Applicability to Dry Storage and Transportation

................................... 2-14 38 2.3.4.1 Use of Static Test Results to Evaluate Safety Margins 39 i n a n HAC Side Drop Event

.................................................. 2-17 40 2.3.4.2 Dynamic Response o f a Fuel Rod ........................................ 2-19 41 2.3.4.3 Seismic Response o f a Fuel Rod .......................................... 2-19 42 2.4 Application of Fatigue Test Results

................................................................. 2-20 43 vi 2.4.1 Lower Bound Fatigue S-N Curves ....................................................... 2-20 1 2.4.2 Fatigue Cumulative Damage Model

..................................................... 2-22 2 2.4.3 Applicability to Storage and Transportation

......................................... 2-23 3 3 DRY STORAGE OF HIGH BURNUP SPENT NUCLEAR FUEL ...................... 3-1 4 3.1 Introduction

.......................................................................................................3-1 5 3.2 Uncanned Fuel (Intact and Undamaged Fuel)

...................................................3-4 6 3.2.1 Leaktight Confinement

...........................................................................3-6 7 3.2.2 Non-Leaktight Confinement

...................................................................3-7 8 3.2.3 Dry Storage U p To 20 Years

............................................................... 3-10 9 3.2.4 Dry Storage Beyond 20 Years

............................................................. 3-11 10 3.2.4.1 Supplemental Results from Confirmatory 11 Demonstration

...................................................................... 3-11 12 3.2.4.1.1 Initial Licensing or Certification

........................... 3-12 13 3.2.4.1.2 Renewal Applications

......................................... 3-12 14 3.2.4.2 Supplemental Safety Analyses

............................................. 3-12 15 3.2.4.2.1 Materials and Structural

...................................... 3-13 16 3.2.4.2.2 Confinement

....................................................... 3-13 17 3.2.4.2.3 Thermal .............................................................. 3-14 18 3.2.4.2.4 Criticality

............................................................. 3-15 19 3.2.4.2.5 Shielding ............................................................. 3-17 20 3.3 Canned Fuel (Damaged Fuel) ......................................................................... 3-19 21 4 TRANSPORTATION OF HIGH BURNUP SPENT NUCLEAR FUEL

............... 4-1 22 4.1 Introduction

.......................................................................................................4-1 23 4.2 Uncanned Fuel (Intact and Undamaged Fuel)

...................................................4-4 24 4.2.1 Leaktight Containment

...........................................................................4-7 25 4.2.2 Non-Leaktight Containment

...................................................................4-7 26 4.2.3 Direct Shipment from the Spent Fuel Pool and Shipment o f 27 Previously Dry

-Stored Fuel (U p To 20 Years since Fuel was 28 Initially Loaded)

................................................................................... 4-11 29 4.2.4 Shipment of Previously Dry

-Stored Fuel (Beyond 20 Years 30 since Fuel was Initially Loaded) ........................................................... 4-12 31 4.2.4.1 Supplemental Data from Confirmatory Demonstration

.......... 4-12 32 4.2.4.2 Supplemental Safety Analyses

............................................. 4-12 33 4.2.4.2.1 Materials and structural

...................................... 4-13 34 4.2.4.2.2 Containment

....................................................... 4-13 35 4.2.4.2.3 Thermal .............................................................. 4-14 36 4.2.4.2.4 Criticality

............................................................. 4-15 37 4.2.4.2.5 Shielding ............................................................. 4-18 38 4.3 Canned F uel ................................................................................................... 4-20 39 5 CONCLUSIONS

................................................................................................ 5-1 40 6 REFERENCES

.................................................................................................. 6-1 41 7 GLOSSARY ...................................................................................................... 7-1 42 vii LIST OF FIGURES 1 Figure 1-1 Hydride Content [H] and Distribution (Average , Inner 2/3 Diameter Of 2 Cladding) i n HBU SNF Cladding (From Billone et al., 2013) .............................. 1-6 3 Figure 1-2 Dissolution (C d) and Precipitation (C p) Concentration Curves Based o n t he 4 Data o f Kammenzind e t al. (1996) for Non-Irradiated Zircaloy

-4 (Zry-4) ............ 1-9 5 Figure 1-3 Publicly-Available Data Collected b y EPRI For PWR End-Of-Life Rod 6 Internal Pressures at 25°C................................................................................. 1-12 7 Figure 1-4 Calculated Rod Internal Pressures for the First 10 Cycles o f t he Watts 8 Bar Nuclear Plant Unit 1 Reactor Under Vacuum Drying Conditions

............... 1-13 9 Figure 1-5 Aggregated Measured and Calculated Values for End-Of-Life Rod Internal 10 Pressures for PWR Fuel Rods. Pressures are Presented at 25 °C (77 °F) .... 1-14 11 Figure 1-6 Calculated Rod Internal Pressure a s a Function o f the Spatially Averaged 12 Gas Temperature for PWR Fuel Rods (i.e. Standard Rods), ZIRLO-Clad 13 IFBA Rods With Hollow (Annular) Blanket Pellets, a nd ZIRLO-Clad IFBA 14 Rods with Solid Blanket Pellets

......................................................................... 1-15 15 Figure 1-7 Fuel Cladding Tube with Stress Element

.......................................................... 1-16 16 Figure 1-8 Calculated Values of Cladding Hoop Stress Vs.

the Spatially

-Averaged 17 Internal Gas Temperature for Standard 17x17 PWR Fuel Rods with 18 Corrosion Layers of 10 µm, 40 µm, and 80 µm................................................. 1-17 19 Figure 1-9 PWR Hoop Stress a s a Function of Internal Gas Temperature for 17 x 17 20 IFBA Fuel Rods (for Both Hollow Blanket and Solid Blanket Pellets) with a 21 40-µm Corrosion Layer

...................................................................................... 1-18 22 Figure 1-10 RCT o f a Sectioned Cladding Ring Specimen i n ANL's Instron's 23 8511 Test Setup

................................................................................................. 1-19 24 Figure 1-11 Ductility Vs. RCT for Two PWR Cladding Alloys Following Slow Cooling 25 from 400°C at Peak Target Hoop Stresses of 110 Mpa and 140 Mpa

............. 1-20 26 Figure 1-12 Ductility Data, as Measured b y RCT , for As-Irradiated Zircaloy

-4 a nd 27 Zircaloy-4 Following Cooling from 400 °C Under Decreasing Internal 28 Pressure and Hoop Stress Conditions

.............................................................. 1-21 29 Figure 1-13 Ductility Data, as Measured b y RCT , for As-Irradiated ZIRLO a nd ZIRLO 30 Following Cooling from 400 °C Under Decreasing Internal Pressure a nd 31 Hoop Stress Conditions

..................................................................................... 1-22 32 Figure 1-14 Ductility Data, as Measured By RCT , for As-Irradiated M5 a nd M5 33 Following Cooling from 400 °C Under Decreasing Internal Pressure a nd 34 Hoop Stress Conditions

..................................................................................... 1-23 35 Figure 1-15 Geometric Models for Spent Fuel Assemblies i n Transportation Packages .... 1-25 36 Figure 2-1 Horizontal Layout o f ORNL U-Frame Setup, Rod Specimen and Three 37 Lvdts for Curvature Measurement, and Front View of CIRFT Installed i n 38 ORNL Hot Cell ..................................................................................................... 2-2 39 Figure 2-2 Schematic Diagram of End and Side Drop Accident Scenarios

......................... 2-5 40 Figure 2-3 Typical Composite Construction o f a Bridge ....................................................... 2-6 41 viii Figure 2-4 Influence o f cg Position on Composite Beam Stiffness

...................................... 2-7 1 Figure 2-5 Images of Cladding

-Pellet Structure i n HBU SNF Rod ....................................... 2-8 2 Figure 2-6 Approximate Extreme Fiber Tensile Stresses Between Pellet

-Pellet Crack

...... 2-9 3 Figure 2-7 Comparison of CIRFT Static Bending Results with Calculated PNNL 4 Moment Curvature (Flexural Rigidity) Derived from Cladding-Only 5 Stress-Strain Curve

............................................................................................ 2-10 6 Figure 2-8 Characteristic Points on Moment-Curvature Curve

.......................................... 2-11 7 Figure 2-9 High Magnification Micrograph Showing Radial Hydrides o f a HBR HBU 8 SNF Hydride-Reoriented Specimen Tested Under Phase II (Specimen 9 HR1 Results Shown; Hydrogen Content 360-400 Wpp m) ............................ 2-15 10 Figure 2-10 Representative Conditions Used for Radial Hydride Treatment for 11 Preparation o f HBR HBU SNF Hydride-Reoriented Specimens Tested 12 Under Phase II ................................................................................................... 2-16 13 Figure 2-11 Plots of Half o f the Cladding Strain Range

(/2) a nd the Maximum Strain 14 (//max) a s a Function of Number of Cycles to Failure........................................ 2-21 15 Figure 2-12 CIRFT Dymanic (Fatigue) Test Results for As-Irradiated a nd 16 Hydride-Reoriented H.B. Robinson Zircaloy

-4 HBU Fuel Rods........................ 2-22 17 Figure 3-1 Example Licensing and Certification Approaches for Dry Storage of High 18 Burnup Spent Nuclear Fuel

................................................................................. 3-3 19 Figure 3-2 First Approach for Evaluating Design

-Bases Drop Accidents During 20 Dry Storage

.......................................................................................................... 3-5 21 Figure 3-3 Second Approach for Evaluati on of Design-Bases Drop Accidents During 22 Dry Storage

.......................................................................................................... 3-6 23 Figure 4-1 Example Approaches for Approval of Transportation Packages with High 24 Burnup Spent Nuclear Fuel

................................................................................. 4-3 25 Figure 4-2 First Approach for Evaluation of Drop Accidents During Transport

.................... 4-5 26 Figure 4-3 Second Approach for Evaluation of Drop Accidents During Transport

.............. 4-6 27 Figure 4-4 Evaluation of Vibration Normally Incident to Transport

....................................... 4-7 28 ix LIST OF TABLES 1 Table 2-1 Comparison of Average Flexural Rigidity Results between CIRFT Static 2 Testing a nd PNNL Cladding-Only Data............................................................. 2-12 3 Table 2-2 Characteristic Points and Quantities Based on Moment-Curvature Curves

..... 2-12 4 Table 2-3 Properties of PWR 15 x 15 SNF R od ................................................................ 2-17 5 Table 2-4 Summary of CIRFT Dynamic Test Results for As-Irradiated and Hydride

-6 Reoriented HBR HBU SNF

................................................................................ 2-20 7 Table 2-5 Coordinates for Lower

-Bound Enveloping S

-N Curve for the HBR HBU 8 SNF Rods........................................................................................................... 2-21 9 Table 3-1 Fractions of Radioactive Materials Available for Release from HBU SNF 10 Under Conditions o f Dry Storage

......................................................................... 3-8 11 Table 4-1 Fractions of Radioactive Materials Available for Release from HBU SNF 12 Under Conditions Of Transport

............................................................................ 4-9 13

xi ACKNOWLEDGMENTS 1 The working group is very grateful to M. Billone (Argonne National Laboratory) for providing 2 valuable input for the writing of the report. R. Einziger (Nuclear Waste Technical Review Board), 3 J. Wang (Oak Ridge National Laboratory), and M. Flanagan

-Bales (U.S. Nuclear Regulatory4 Commission) also provided valuable insights, observations, and recommendations.

5

xiii ABBREVIATIONS AND ACRONYMS 1 ADAMS Agencywide Documents Access and Management System AMP aging management program ANL Argonne National Laboratory ANS American Nuclear Society ANSI American National Standards Institute b width BWR boiling-water reactor C d concentration at d is solution C p concentration at precipitation CFR Code of Federal Regulations cg center of gravity CoC Certificate of Compliance CIRFT cyclic integrated reversible

-bending fatigue tester CRUD C h a lk River unknown deposit CWSR A cold worked stress relieved annealed p/D mo offset strain dp temperature hysteresis (dissolution

-precipitation

) D mi inner (metal) cladding diameter D mo outer (metal) cladding diameter DLF dynamic load factor DTT ductility transition temperature DOE U.S. Department of Energy DSS dry storage system average tensile strain

-N strain per number of cycles E elastic modulus E c elastic modulus of the cladding E p elastic modulus of the fuel pellet EOL end-of-life EPRI Electric Power Research Institute GBC general burnup credit GTCC greater-than-Class-C waste h height h m cladding (metal) thickness HAC hypothethical accident conditions (transportation)

HBR H. B. Robinson HBU high burnup HRT hydride reorientation treatment Hz hertz I moment of inertia I c moment of inertia of the cladding

xiv I p moment of inertia of the fuel pellet IAEA International Atomic Energy Agency IFBA integral fuel burnable absorber ISFSI independent spent fuel storage installation ISG Interim Staff Guidance curvature -N curvature per number of cycles keff k-effective l rod length between spacers LBU low burnup LVDT linear variable differential transformer M bending moment n i number of strain cycles at strain level i N i number of strain cycles to produce failure at i NCT normal conditions of transport NRC U.S. Nuclear Regulatory Commission ORNL Oak Ridge National Laboratory P i rod internal pressure P o rod external pressure PNNL Pacific Northwest National Laboratory PWR pressurized-water reactor r outer radius RCT ring compression testing RHCF radial hydride continuity factor RIP rod internal pressure RXA recrystallized annealed average tensile stress cladding hoop stress z cladding longitudinal stress SNF spent nuclear fuel SRP standard review plan SSC structure, system, and component T d dissolution temperature T p precipitation temperature w uniform applied load ymax distance to the neutral axis

xv Units of Measure C Celsius F Fahrenheit ft foot g 9.806 m/s 2 GWd/MTU gigawatt-days per metric ton of uranium h hour i n. inch lb pound m meter micrometer, 1 x 10-6 meter mm millimeter, 0.001 meter MPa megapascal, 1 x 10 6 pascals N newton N*m newton meter Pa pascal psi pounds per square inch s second Torr Torr (unit of pressure) wppm parts per million by weight

1-1 1 1 1.1 2 As required by Title 10 of the Code of Federal Regulations (10 CFR) 72.44(c), a specific license 3 for dry storage of spent nuclear fuel (SNF) is to include technical specifications that, among 4 other things, define limits on the fuel and allowable geometric arrangements. , Further, a s 5 required by 10 CFR 72.236(a), a Certificate of Compliance (CoC) for a dry storage system 6 (DSS) design must include specifications for the type of spent fuel (i.e., boiling water reactor 7 (BWR), pressurized water reactor (PWR), or both), maximum allowable enrichment of the fuel 8 prior to any irradiation, burn

-up (i.e., megawatt

-days/MTU), minimum acceptable cooling time of 9 the spent fuel before storage in the spent fuel storage cask, maximum heat designed to be 10 dissipated, maximum spent fuel loading limit, condition of the spent fuel (i.e., intact assembly or 11 consolidated fuel rods), and inerting atmosphere requirements. These specifications ensure 12 that the loaded SNF assemblies remain within the bounds of the safety analyses in the 13 approved design basis.

14 The regulations in 10 CFR Part 72, "Licensing Requirements for the Independent Storage of 15 Spent Nuclear Fuel, High

-Level Radioactive Waste, and Reactor

-Related Greater Than Class C 16 Waste," include a number of fuel

-specific and DSS

-specific requirements that may be 17 dependent of the design-basis condition of the fuel cladding. As required by 10 CFR 18 72.122(h)(1), the SNF cladding is to be protected against degradation that leads to gross 19 ruptures or the fuel must be otherwise confined such that degradation of the fuel during storage 20 will not pose operational safety problems with respect to its removal from storage. In addition, 21 10 CFR 72.122(l) states that the DSS must be designed to allow ready retrieval of the SNF. 22 According to Interim Staff Guidance (ISG)

-2, Revision 2, "Fuel Retrievability in Spent Fuel 23 Storage Applications,"

issued in April 2016 (NRC, 2016a).1 This may be demonstratedon an 24 assembly basis per the approved design basis. The condition of the fuel cladding may also 25 impact the safety analyses used to demonstrate compliance with DSS

-specific requirements in 26 10 CFR 72.124(a), 10 CFR 72.128, and 10 CFR 72.236(m).

27 Similarly for transportation, the regulations in 10 CFR Part 71, "Packaging and Transportation of 28 Radioactive Material," also include a number of fuel

-specific and package

-specific requirements.

29 The regulations in 10 CFR 71.31 , "Contents of Application" and 10 CFR 71.33, "Package 30 description," requires an application for a transportation package to describe the proposed 31 package in sufficient detail to identify the package accurately and provide a sufficient basis for 32 evaluation of the package, which includes a description of the chemical and physical form of the 33 allowable contents. The regulations in 10 CFR Part 71 also require that (1) the geometric form 34 of the package contents not be substantially altered under the tests for normal conditions of 35 transport (NCT) (10 CFR 71.55(d)(2))

and (2) a package used for the shipment of fissile material 36 is to be designed and constructed and its contents so limited that under the tests for 37 hypothetical accident conditions (HAC) specified in 10 CFR 71.73, "Hypothetical Accident 38 Conditions," the package remains subcritical (10 CFR 71.55(e). The requirement assumes that 39 1 The current revisions of all ISG documents will be rolled into revised standard review plans (SRPs) for dry storage and transportation of SNF, as appropriate, and will then be removed from the public domain. The revised SRPs will be issued for public comment prior to being finalized.

1-2the fissile material is in the most reactive credible configuration consistent with the 1 damaged condition of the package and the chemical and physical form of the contents 2 (10 CFR 71.55(e)(1)).

3 To comply with the requirements mentioned above, the fuel cladding generally serves a design 4 function in both DSSs and transportation packages for ensuring that the configuration of 5 undamaged and intact fuel remains within the bounds of the reviewed safety analyses.

2 6 Therefore, an application should address potential degradation mechanisms that could result in 7 gross cladding ruptures during operations. To assist the safety review of potential degradation 8 mechanisms, the U.S. Nuclear Regulatory Commission (NRC) staff (the staff) has historically 9 issued guidance on acceptable storage and transport conditions that limit SNF degradation 10 during operations and ensure that the reviewed safety analyses remain valid.

11 1.2 Fuel Cladding Performance and Staff's Review Guidance 12 Time-dependent (i.e., age

-related, not event

-related) mechanisms resulting in changes to the 13 fuel cladding performance are all primarily driven by the fuel's temperature, rod internal 14 pressure (and corresponding pressure

-induced cladding hoop stresses), and the environment 15 during dry storage or transport operations

. Contrary to the hoop stresses experienced by t he 16 fuel cladding during reactor operation, which are generally compressive because of the high 17 reactor coolant pressure , t he hoop stresses during drying

-transfer, dry storage, and transport 18 operations are tensile because of the low pressure external to the cladding. For instance, the 19 pressure of the environm ent surrounding the fuel in the reactor can be 16 MPa (1.2 x 10 5 Torr) 20 while the environment surrounding the fuel in the DSS confinement cavity may be as low as 21 400 Pa (3 Torr) at the end of vacuum drying and 0.5 MPa (3.75 x 10 3 Torr) during dry storage.

22 The magnitude of the cladding hoop stress es will depend on the differential pressure across the 23 cladding wall and thus the rod internal pressure at a given time. Various factors determine the 24 rod internal pressu re , including the fuel's fabrication and irradiation conditions (i.e., fabrication 25 gas fill pressure, cladding thickness, presence of burnable absorbers, burnup) and the average 26 gas temperature within the fuel rods.

The average gas temperature within the fuel rods has a 27 first-order effect on the hoop stress in the cladding and thus cladding performance, and 28 therefore it is critical to controlling the peak cladding temperature of the fuel rods during vacuum 29 drying and storage/transport operations to temperatures demonstrated to preserve cladding 30 integrity. 31 To assist in the safety review of DSS and transportation packages, the staff has developed 32 guidance with a supporting technical bas is for setting adequate fuel conditions, including 33 acceptable peak cladding temperatures during short

-term loading operations so that the 34 cladding meets the pertinent regulations. Historically, guidance has been issued as ISG-11, 35 "Cladding Considerations for the Transportation and Storage of Spent Fuel,"

which has b een 36 revised multiple times to incorporate new data and lessons learned from the staff's review 37 experience. Initial standard review plans (SRPs) prior to ISG

-11 stated that DSSs and 38 transportation packages needed to be dried to a level where galvanic corrosion could be ruled 39 out as a fuel degradation mechanism. The guidance specified moisture levels only for low 40 burnup (LBU) fuel (i.e., burnup below 45 GWd/MTU) because of the lack of degradation data at 41 higher burnup values. In 1999, the staff first issue d ISG-11 to supplement the SRPs by 42 2 If the fuel is classified as damaged, a separate canister (e.g., a can for damaged fuel) that confines the assembly contents to a known volume may be used to provide this assurance.

1-3addressing potential degradation of high burnup (HBU) fuel (i.e., burnup exceeding 1 45 GWd/MTU). 2 A year later, the staff issued ISG-11, Revision 1 to incorporate new data, but also to give the 3 applicant the responsibility for demonstrating that the cladding was adequately protected. ISG-4 11, Revision 1 stated that cladding oxidation should not be credited as load

-bearing in the fuel 5 cladding structural evaluation and also defined a 1

-percent creep strain limit on the cladding.

It 6 also discussed the use of damaged fuel cans for confining fuel with gross ruptures. ISG

-11, 7 Revision 1, accounted only for Zircaloy

-clad fuel rods and not for other advanced cladding alloys 8 (e.g., ZIRLO and M5). 9 In 2002, the staff issued ISG-11, Revision 2, to change the definition of damaged fuel, remove 10 the 1-percent creep strain limit, and discuss criteria to limit hydride reorientation in the cladding.

11 It also made the guidance applicable to all zirconium

-based claddings and all burnup levels.

12 The revision described onerous calculations, dependent on the characteristics of the fuel to be 13 stored, to determine the maximum cladding temperature for the design

-bas is fuel per a justified 14 creep strain limit.

Gruss et al. (2004) discuss in more detail the data used for supporting ISG-15 11, Revision 2. Historically, ISG-11 has not discussed the use of an inert atmosphere to 16 mitigate fuel degradation. Peehs (1998) indicated that air could be used as an atmosphere 17 below 200 °C (392 °F) but later research indicated a lower temperature was necessary.

18 Therefore, ISG-22, "Potential Rod Splitting Due to Exposure to an Oxidizing Atmosphere during 19 Short-Term Cask Loading Operations in LWR or Other Uranium Oxide Based Fuel," issued 20 May 2006 (NRC, 2006), addressed the use of an inert atmosphere for loading operations. 21 In November 2003, the staff issued ISG-11, Revision 3 , "Cladding Considerations for the 22 Transportation and Storage of Spent Fuel

" (NRC, 2003a). The guidance was eventually 23 incorporated into NUREG

-1536, Revision 1, "Standard Review Plan for Spent Fuel Dry Storage 24 Systems at a General License Facility," issued in July 2010 (NRC, 2010), although not yet 25 incorporated into a revision of NUREG

-1567, "Standard Review Plan for Spent Fuel Storage 26 Facilities," issued in March 2000 (NRC, 2000 a) (i.e. the standard review plan for specific 27 licenses under 10 CFR Part 72). ISG-11, Revision 3 replaced the calculation of the maximum 28 cladding temperature per a justified creep strain limit with a generic 400 °C (752 °F) peak 29 cladding temperature limit applicable to normal conditions of storage and transportation, as well 30 as short-term loading operations (e.g., drying, backfilling with inert gas, and transfer of the DSS 31 cask or canister to the storage pad). ISG

-11, Revision 3 also defined a higher short

-term 32 temperature limit applicable to LBU fuel if the applicant demonstrated by calculation that the 33 cladding hoop stress would not exceed 90 MPa (1.3 x 10 4 psi) for the proposed temperature 34 limit. The guidance also defined a generic maximum cladding temperature limit of 570 °C 35 (1,058 °F) for off

-normal and accident conditions applicable to all burnups.

36 I n additi on to creep, I SG-1 1 Revision 3 (NRC, 2003 a), also considered minimizing hy dr ide 37 reorientation. At the time of its issuance, the technical basis discussed in ISG

-11, Revision 3 38 supported the staff's conclusion that hydride reorientation would be minimized by maintaining 39 cladding temperatures below 400 °C (752 °F) and restricting the change in cladding 40 temperatures during drying

-transfer operations to less than 65 °C (117 °F). Th is temperature 41 change limit was based on the temperature drop required to obtain the degree of 42 supersaturation required for the precipitation of radial hydrides in a short thermal cycle 43 (see Section 1.5

.1). Therefore, ISG

-11, Revision 3, states that the cladding should not 44 experience more than 10 thermal cycles, each not exceeding 65 °C (117 °F), which provided 45 assurance that hydride reorientation would be limited.

46 1-4Research results obtained since the ISG

-11, Revision 3, have shown that hydride reorientation 1 c an still occur below the generic 400 °C (752 °F) peak cladding temperature limit (Aomi et al, 2 2008; Billone et al., 2013; Billone et al., 2014; Billone et al., 2015

). To better understand 3 hydride reorientation, both the NRC and the U.S. Department of Energy (DOE) have obtain ed 4 additional data on the performance of HBU SNF cladding with reoriented hydrides to determine 5 if the guidance in ISG

-11, Revision 3, ought to be revised. 6 1.3 Cladding Creep 7 Creep is the time-dependent deformation of a material under stress. The main driving force for 8 cladding creep at a given temperature is the hoop stress caused by internal rod pressure, which 9 results from the fission and decay gases released to the gap between the fuel and cladding (Ito, 10 at al., 2004). Fuel pellet swelling may also result in localized stresses on the cladding due to 11 the mechanical interaction between the cladding and the fuel. Pellet swelling may occur due to

12 (1) the incorporation of soluble and insoluble solid fission products in the fuel matrix, (2) the 13 formation of intra- and intergranular fission gas bubbles, particularly in the hot interior region of 14 a fuel pellet, and (3) the formation of a large number of small gas bubbles in the fine

-grained 15 ceramic structure that builds inward from the outer pellet surface for HBU fuel. If excessive 16 creep of the cladding were to occur during dry storage, it could lead to thinning, hairline cracks, 17 or gross ruptures (Hanson et al, 2012) and potentially compromise the ability to safely retrieve 18 by normal means the HBU fuel on a single

-assembly basis (if required by the design basis). 19 The appendix to ISG-11, Revision 3 (NRC, 2003a) reviewed the data used by the staff to obtai n 20 reasonable assurance that creep will not result in gross ruptures for peak cladding temperatures 21 below 400 °C (752 °F). The fabrication of fuel rods is such that the creep of the cladding is self

-22 limiting. As the cladding creeps, the internal volume of the rod increases and stress decreases.

23 However, as the gas volume within the fuel column increases, the average gas temperature 24 also increases

. The net effect is a slow decrease in pressure and hoop stress with increasing 25 creep strain. The stress also decreases with increasing storage or transport time due to the 26 decrease in internal pressure with decreasing temperature. ISG

-11, Revision 3, concluded th e 27 following:

28 1.deformation caused by creep will proceed slowly over time and will decrease the rod 29 pressure, 30 2.the decreasing cladding temperature also decreases the hoop stress, and this too will 31 slow the creep rate so that during later stages of dry storage, further creep deformation 32 will become exceedingly small, and 33 3.in the unlikely event that a breach of the cladding due to creep occurs, it is believed that 34 this will not result in gross rupture.

35 These conclusions are considered applicable to fuel at all burnups because the relatively small 36 differences in creep rate as a function of materials and burnup are not expected to have a 37 significant impact on the maximum creep strains in the rod. The technical basis in ISG

-11, 38 Revision 3 (NRC, 2003a) has provided reasonable assurance to the staff that creep strains 39 during dry storage and transportation will not result in fuel failures nor compromise the assumed 40 fuel configuration in the safety analyses. However, the staff recognizes the uncertainties 41 associated with extrapoling short-term accelerated test data to extended periods of dry storage.

42 The staff further recognizes the separate

-effects nature of the accelerated creep testing 43 conducted to date, which would not account for potential combined effects with other 44 phenomena occurring during dry storage (e.g., annealing of irradiation hardening, hydride 45 reorientation). Therefore, the staff considers it prudent that long

-term observation of HBU SNF 46 1-5stored in a deployed DSS be used to confirm the conclusions of the accelerated short

-term 1 testing. To aid users in demonstrating adequate creep performance during storage periods 2 beyond 20 years, in June 2016, the staff issued guidance in NUREG-1927, Revision 1 , 3 "Standard Review Plan for Renewal of Specific Licenses and Certificates of Compliance for Dry 4 Storage of Spent Nuclear Fuel" (NRC, 2016b), which discusses the use of an Aging 5 Management Program using a surrogate surveillance and monitoring program to provide this 6 confirmatory long term data.

7 1.4 Effects of Hydrogen o n Cladding Mechanical Performance 8 During irradiation, hydrogen is generated due to water

-coolant corrosion (i.e., oxidation) of the 9 cladding, which diffuses into the zirconium

-based material. As the solubility limit of hydrogen in 10 the cladding is exceeded, circumferential hydrides precipitate (Figure 1-1). The preferential 11 circumferential precipitation of the hydrides during reactor operation results from the texture of 12 cladding, which is determined by the manufacturing process. The number density of these 13 circumferential hydrides varies across the cladding wall due to the temperature drop from the 14 fuel side (hotter) to the coolant side (cooler) of the cladding. When the cladding absorbs 15 significant hydrogen, migration and precipitation of dissolved hydrogen into the coolant side of 16 the cladding can result in the formation of a rather dense hydride rim just below the outer 17 cladding oxide layer. The hydride number density and thickness of this hydride rim depend on 18 cladding design and reactor operating conditions for a given fuel type. For example, fuel rods 19 operated at high linear heat rating to high burnup generally have a very dense hydride rim that 20 is less than 10 percent of the cladding wall thickness.

Conversely, fuel rods operated at low 21 linear heat ratings to high burnup have a more diffuse hydride distribution that could extend as 22 far as 50 percent across the cladding wall.

23 1-6 Figure 1-1 Hydride Content [H] and Distribution (Average, Inner 2/3 Diameter of 1 Cladding) i n HBU SNF Cladding (from Billone et al., 2013) 2 The staff concluded in ISG-11, Rev ision 3 (NRC, 2003a), that the hydride rim, along with any 3 cladding metal oxidized during reactor operation, should not be considered as load bearing 4 when determining the effective cladding thickness for the structural evaluation of the assembly 5 in the DSS or transportation package. However, the staff recognizes that there is no reliable 6 predictive tool available to calculate this rim thickness, which varies along the fuel

-rod length, 7 around the circumference at any particular axial location, from fuel rod to fuel rod within an 8 assembly, and from assembly to assembly. Moreover, recent data generated by Argonne 9 National Laboratory (ANL) have shown that for the full range of gas pressures anticipated during 10 drying and storage, the hydride rim remains intact following slow cooling under conditions of 11 decreasing pressure. The results suggest that hydride rims have some load bearing capacity 12 (Billone et al., 2013; Billone et al., 2014; Billone et al., 2015). These results indicate that it may 13 be appropriate to include the hydride rim in the effective cladding thickness calculation. 14 Therefore, the staff considers acceptable the inclusion of the hydride rim thickness in the 15 calculation of the effective cladding thickness when mechanical test data referenced in the 16 structural evaluation ha ve adequately accounted for its presence. Historically, this has been the 17 case during the review of DSS and transportation packages, as applicants have provided 18 1-7mechanical property data generated from tests with irradiated cladding samples with an intact 1 hydride rim. These data includes test results derived from uniaxial tensile tests or pressurized 2 tube tests of samples that do not have a machined gauge section.

3 Applicants have generally relied on a public database of materials properties for Zircaloy

-4, 4 Zircaloy-2 and ZIRLO to analyze the behavior of as

-irradiated cladding (Geelhood et al, 2008; 5 Geelhood et al, 2013) during dry storage and transportation. Additional data for engineering 6 properties (e.g., yield stress, ultimate tensile stress, and uniform elongation) can be found in the 7 open literature for ZIRLO (Cazalis et al

., 2005; Pan et al

., 2013), Optimized ZIRLO (Pan et al., 8 2013), and M5 (Cazalis et al

., 2005; Fourgeaud et al., 2009; Bouffioux et al

., 2013). The 9 applicant should adequately justify the use of any of these properties for the fuel designs cited 10 for use in the DSS or transportation package application. Any use of mechanical properties 11 from uniaxial

-tension and ring

-expansion tests on cladding specimens with machined gauge 12 sections, where some of the hydride rim would have been inadvertently removed during outer

-13 surface oxide removal, should be adequately justified. The mechanical property data from 14 these specimens are still valuable, but characterization of t he ir remaining rim thickness, post-15 test determination of their hydrogen concentration, or both may be needed. 16 1.5 Hydride Reorientation 17 As discussed in Section 1.4, hydrogen infiltrates the cladding during reactor operation. The 18 excess hydrogen (i.e., hydrogen exceeding the solubility limit in the cladding) precipitates 19 primarily in the circumferential

-axial direction. However, under temperature and stress 20 conditions experienced during vacuum drying and storage/transport operations, some of these 21 hydrides may redissolve and subsequently reprecipitate as new hydrides. During this process, 22 the orientation of these precipitated hydrides may change from the circumferential

-axial to the 23 radial-axial direction. 24 The technical basis discussed in ISG-11, Revision 3 (NRC, 2003a) has supported the staff's 25 conclusion that if peak cladding temperatures are maintained below 400 °C (752 °F) and the 26 pressure-induced hoop stresses in the cladding were maintained below 90 MPa (1.3 x 10 4 psi), 27 then hydride reorientation would be minimized. The database used for this determination (see 28 Figure 3 in Chung, 2004) had a mixture of results from irradiated and non

-irradiated material, 29 high and low hydrogen concentrations, different cladding types, different cooling rates, and 30 other variables. In addition, the methods to determine if there were radial hydrides varied 31 considerably from researcher to researcher. Since the issuance of ISG

-11, Revision 3, 32 research results generated at ANL (Billone et al

., 2013; Billone et al., 2014; Billone et al., 2015) 33 and in Japan (Aomi et al

., 2008) have shown that hydride reorientation can still occur at lower 34 temperatures and stresses than those assumed in ISG

-11, Revision 3. Because of the number 35 of variables involved, the staff agreed that it would not be practical to precisely determine the 36 temperature and stress conditions to prevent reorientation. Rather, t he critical question wa s 37 what effect hydride reorientation would ha ve on the mechanical behavior of the cladding , 38 particularly since the desi gn-bas is structural evaluation of the SNF assembly generally assumes 39 as-irradiated cladding mechanical properties (i.e., properties not accounting for hydride 40 reorientation). If hydride reorientation had an observable effect on the mechanical behavior of 41 the cladding (i.e., it decreased the failure strain limit of the cladding in response to stresses 42 during operations), then the failure limits as defined in the design

-bas is structural evaluations 43 would have to be modified. 44 Because both circumferential and radial hydrides are oriented in the planes parallel to the 45 principal normal tensile stress during bending loading, the staff has expected that HBU SNF 46 1-8fatigue strength and bending stiffness would not be sensitive tothe hydride orientation under1 bendi ng mo ments t hat produc e longitudinal tensil e stresses i n t he rod (T ang et al , 2015)3 2 Experimental confirmation of this expectati on was prudent. Therefore, the NRC and DOE 3 conducted complementary research programs to investigate the cyclic fatigue and bend in g 4 strength performance of HBU SNF cladding in both as-irradiated and reoriented conditions 5 (Wang et al

., 201 6; NRC, 2017). 6 Even with the expectation that hydride orientation would not have a significant impact on the 7 fatigue strength and bending stiffness of HBU SNF under bending moments that produce 8 longitudinal tensile stresses in the rod, the staff expressed concern that hydride orientation 9 could impact the failure stresses and strains under pinch

-type loads. Pinch

-type loads could 10 potentially occur during postulated drop accidents in storage, normal conditions of transport 11 (NCT), or hypothetical accident conditions (HAC) during transportation. The staff was 12 particularly concerned with reduced cladding ductility during the HAC 9-m (30-ft) side drop or a 13 tip-over handling accident, where pinch loads could occur due to rod-to-grid spacer contact, rod-14 to-rod contact, or rod-to-basket contact. If the fuel temperature were to be sufficiently low at the 15 time of the accident, these pinch loads could compromise the analyzed fuel configuration.

16 Thus, research was conducted in the United States and Japan to study the ductility of cladding 17 with reoriented hydrides under diametrically

-opposed pinch loads. Ring compression testing 18 (RCT) was used to assess residual ductility of de-fueled HBU SNF cladding specimens 19 subjected to hydride reorientation (see Section 1.5

.4). Th is testing led to the establishment of a 20 ductility transition temperature (DTT) (i.e., a temperature at which the tested cladding segments 21 were determined to lose ductility relative to as

-irradiated cladding

). Th e following section 22 discuss important parameters affecting the DTT and provides the staff's conclusion on its 23 relevance for future licensing and certification actions involving HBU SNF. 24 1.5.1 Hydride Dissolution and Precipitation 25 During drying

-transfer operations, the cladding temperature increases, which causes some of 26 the circumferential hydrides to dissolve as hydrogen. The amount of hydrogen dissolved 27 depends on the temperature (T d) and increases according to the solubility curve (C d) for 28 zirconium-based alloys (Kammenzind et al

., 1996; Kearns, 1967; McMinn, et al

., 2000).

29 Zirconium-based alloys are materials that can have hydrogen in a supersaturated solution 30 because of the extra energy (strain, thermal) required to precipitate zirconium hydrides in the 31 cladding matrix. This results in a hysteresis in the solubility

-precipitation curves as shown in 32 Figure 1-2. 33 3Hydrides are essentially two

-dimensional features since their thickness is relatively small compared to the other two dimensions. Radial hydrides span in the longitudinal and radial directions, and circumferential hydrides span in the longitudinal and circumferential directions. The bending tensile stresses are in the longitudinal direction.

Therefore, the bending tensile stresses are parallel to the plane of both the radial and circumferential hydrides.

1-9Figure 1-2 Dissolution (C d) and Precipitation (C p) Concentration Curves Based on the 1 Data of Kammenzind et al. (1996) for Non-Irradiated Zircaloy-4 (Zry-4) 2 (Revised Figure 1 from Billone, et al., 2014). Also Shown I s the Best Fit t o 3 the Dissolution Curve (C d) for Zirconium (Zr), Zircaloy-2 (Zry-2), a nd 4 Zircaloy-4, Which Includes the Zircaloy

-2 and Zircaloy

-4 Data Generated by 5 Kearns (1967). t dp = T d - T p Refers to the Temperature Drop Required for 6 Precipitation, where T d and T p are the Corresponding Temperatures Iin the 7 Solubility and Precipitation Curves for the Same Hydrogen Content 8 The solubility curves (C d) plotted in Figure 1-2 indicate that the amount of hydrogen that 9 dissolves increases with increasing temperature, but it is relatively independent of alloy 10 composition and fabricated microstructure (recrystallized annealed (RXA) and cold worked 11 stress relieved annealed (CWSRA)) (Kearns, 1967). Both Kammenzind et al (1996) and Kearns 12 (1967) used diffusion couples, with one sample containing excess hydrogen and the other 13 sample containing essentially no hydrogen, exposed to long annealing times (e.g., 2 days at 14 525 °C (977 °F) and 10 days at 260 °C (500 °F)). As shown in Figure 1

-2 , Kearns' dissolution 15 correlation for Zircaloy

-2 and Zircaloy

-4 is in excellent agreement with the correlation of 16 Kammenzind et al. (e.g., 207 wppm versus 210 wppm at 400 °C (752 °F), and 127 wppm versus 17 133 wppm at 350 °C (662 °F)) and is well within experimental error. In terms of precipitation, 18 dp = T d - T p, where T d and T p are the corresponding temperatures in 19 the solubility and precipitation curves at the same hydrogen content) required for precipitation is 20 approximately 65 °C (117 °F). That is, for irradiated cladding that contains no radial hydrides 21 prior to heating, the 65 °C (117 °F) temperature decrease is necessary to initiate precipitation of 22 0 50 100 150 200 250 300 350 400 450 0 100 200 300 400 500Hydrogen Content (wppm)Temperature (°C)Kammenzind Cd (Zry-4)Kammenzind Cp (Zry-4)Kearns Cd (Zr,Zry-2,Zry-4)DissolutionPrecipitationTdp 1-10radial hydrides.

4 However, if circumferential hydrides are present at the peak cladding 1 temperature, some hydrogen will precipitate by growth of the existing circumferential hydrides 2 during this 65 °C (117 °F) temperature drop because of the lower energy required to grow rather 3 than to initiate precipitation of new hydrides (Colas et al

., 2014). The strain field remaining from 4 the regions of the hydrides that dissolved during heating also facilitates the growth of existing 5 hydrides. 6 McMinn et al. (2000) used a different method (differential scanning calorimetry) to generate an 7 independent data set for dissolution

-precipitation curves per non-irradiated and lightly

-irradiated 8 Zircaloy-2 and Zircaloy

-4 samples with low hydrogen content 60 9 wppm) exposed to temperatures less than 320°C (608 °F). The data show the effects of 10 irradiation (increase in solubility), as well as pre

-annealing time and temperature (decrease in 11 solubility). The increase in hydrogen solubility for irradiated materials is likely the result of 12 hydrogen trapped in irradiation

-induced defects. However, it is not clear yet whether the 13 trapped hydrogen is available for precipitation unless the temperature is high enough to anneal 14 out some of these defects. Extrapolation of the dissolution correlation of McMinn et al. (2000) 15 for non-irradiated cladding alloys gives only 172 wppm of dissolved hydrogen at 400 °C (752 °F) 16 and 102 wppm at 350 °C (662 °F), while the data for irradiated cladding agree quite well with 17 the correlations of Kammenzind et al (1996) and Kearns (1967). The staff considers these two 18 sources to be reasonably representative of dry storage and transportation because the long 19 annealing times used to achieve equilibrium for dissolution are more applicable to drying

-20 storage than the much shorter times used for measurements taken by differential scanning 21 calorimetry. Further, the staff considers these data to provide an upper bound for non

-irradiated 22 cladding and close to a best fit for irradiated cladding.

23 The amount of hydrogen dissolved will depend on the peak cladding temperature during drying

-24 transfer, dry storage, and transport operations. This temperature is typically achieved during 25 vacuum drying, which generally takes about 8 to 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> depending on the DSS or transport 26 package design and loading parameters. Figure 1-2, along with an assessment of the axial 27 hydrogen content of the fuel rods and the peak cladding temperature, can be used to estimate 28 the amount of dissolved hydrogen for a given allowable fuel in a DSS or transportation package. 29 The degree of reorientation will depend on the fuel cladding fabrication process, as well as the 30 cladding hoop stresses and temporal thermal profile of the fuel during operations. The following 31 discussions provide additional details on these parameters.

32 1.5.2 Fuel Cladding Fabrication Process 33 The cladding alloy and corresponding fabrication process are important factors for determining 34 the extent of hydride reorientation. Two predominant cladding microstructures are produced 35 during fabrication of zirconium

-based cladding

CWSRA and RXA. Zircaloy

-4 and ZIRLO are 36 generally CWSRA, whereas Zircaloy

-2 and M5 are RXA. Because hydrides tend to precipitate 37 in the grain boundaries, RXA claddings are more susceptible to hydride reorientation, since 38 these cladding types have a larger fraction of grain boundaries in the radial direction (equiaxed 39 grains) relative to CWSRA claddings (which have more elongated grains).

40 4 This hysteresis is the basis for the guidance in ISG

-11, Revision 3 (NRC, 2003a), to limit repeated thermal cycling (repeated heatup/cooldown cycles) during loading operations to less than 10 cycles, with cladding temperature variations that are less than 65

°C (117 °F) each.

1-11 1.5.3 End-Of-Life Rod Internal Pressures and Cladding Hoop Stresses 1 Most rods are initially backfilled with a pressurized inert helium atmosphere to improve thermal 2 conductivity during irradiation and to decrease the rate of cladding creep

-down onto the fuel.

3 During the course of irradiation, fission gases are generated in the fuel pellets.

Some of the 4 fission gas will be released to the void volume within the fuel column and plenum. The fission 5 gas released is about 1 to 3 percent for PWR fuel rods irradiated under low

-to-moderate 6 conditions up to a burnup of about 45 GWd/MTU, at which point the rate of release increases 7 gradually to about 5 to 7 percent for a burnup of 65 GWd/MTU. For BWR fuel rods, the fission 8 gas release can be in the range of 10 to 15 percent at burnups exceeding 45 GWd/MTU. PWR 9 fuel rods with internal burnable poisons (e.g.

, boron-10 in zirconium

-diboride coating on fuel 10 pellets) can also release decay gases (e.g., helium) within the fuel rod. The pressure of these 11 gases in PWR fuel rods increases with burnup due to the increase in fission gas release, the 12 decay gases released from the burnable poisons, and the decrease in void volume resulting 13 from cladding creep down and fuel swelling.

14 The internal pressure of the rod exerts hoop and axial stresses in the cladding, which increase 15 with burnup because of the increase in internal pressure and the decrease in cladding thickness 16 due to waterside corrosion (i.e., oxidation). For BWR fuels, increased cladding oxidation and 17 hydrogen pickup are observed at burnups exceeding 50 GWd/MTU. In PWRs, hydrogen pickup 18 is usually correlated to the oxide thickness, which varies depending on the alloy. The condition 19 of the fuel as it is removed from the reactor is described more fully in the International Atomic 20 Energy Agency (IAEA) Nuclear Energy Series NF

-T-3.8 (IAEA , 2011). 21 Post-irradiation examination of cladding specimens subjected to representative drying

-transfer 22 and dry storage operations has shown that the degree of radial hydride precipitation is very 23 sensitive to the peak cladding hoop stresses. The range of relevant hoop stresses depends on 24 the range of end

-of-life (EOL) rod internal pressures (RIPs), the range of average gas 25 temperatures within fuel rods during drying and storage, and geometric factors used to convert 26 the pressure differences across the cladding to cladding hoop stresses. The following sections 27 discuss these topics.

28 1.5.3.1 End-Of-Life R od Internal Pressures for Pressurized

-Water Reactor Fuel R ods 29 The publicly

-available database for EOL RIPs for PWR fuel rods is sparse relative to the 30 number of rods that have been irradiated. In addition, the RIP data in this database are for 31 standard fuel rods, mostly those clad in zirconium-tin alloy (Zircaloy-4) with older (1975

-1985) 32 fuel designs and reactor operating conditions. Thus, the database is heavily populated with 33 data from what are generally called "legacy" fuel rods. Figure 1-3 shows the publicly

-available 34 data for standard fuel rods, as collected by the Electric Power Research Institute (EPRI) 35 (Machiels, 2013). Data points are for EOL RIP s extrapolated to 25

°C (77 °F), and are identified 36 by the reactor, the assembly design, and the as

-fabricated helium fill pressure at 25°C (77 °F). 37 Data points labeled as "ENUSA" are for fuel rods irradiated in the Vandellos Unit 2 reactor in 38 Spain. 39 The database consists of 92 data points:

40 -4 and 3 ZIRLO) 41 35 in the range of >45 GWd/MTU to 60 GWd/MTU (25 Zircaloy

-4 and 10 ZIRLO) 42 30 in the range of >60 GWd/MTU to 74 GWd/MTU (15 each of Zircaloy

-4 and ZIRLO

) 43 1-12 Publicly available EOL RIP data are not available for M5-clad SNF rods. Helium fill pressures at 1 fabrication range from 2.00 MPa (290 psi) 3.45 MPa (500 psi). The EOL RIP data appear to 2 be relatively flat between about 40 GWd/MTU and 65 GWd/MTU. 3 Figure 1-3 Publicly-Available Data Collected b y EPRI for PWR End-Of-Life 4 Rod Internal Pressures at 25°C (77 °F) (Reproduction of Figure 2-1 from 5 Machiels (2013))

6 Publicly available data are also not available for ZIRLO-clad integral fuel burnable absorber 7 (IFBA) rods (zirconium diboride

-based), which would have the highest EOL RIP values. Given 8 the sparsity of the database and the absence of publicly available data for standard M5

-clad 9 rods and ZIRLO

-clad IFBA rods, predictions are needed for a wide range of advanced cladding 10 alloys, advanced fuel designs, and more current operating conditions.

11 Two more recent public reports provided EOL RIP values for ZIRLO

-clad IFBA rods from 12 calculations performed with FRAPCON, an NRC

-sponsored fuel performance code.

Oak 13 Ridge National Laboratory (ORNL) published a set of calculations for over 68,000 Zircaloy

-4 14 and ZIRLO fuel rods irradiated during the first 10 cycles of the Watts Bar Nuclear Plant Unit 1 15 reactor (Bratton et al, 2015). FRAPCON was used to predict RIPs for standard rods and IFBA 16 rods irradiated for one cycle, two cycles, and three cycles, with each cycle consisting of 17 18 months. ORNL modified the FRAPCON code to account for the helium release from the 18 IFBA rods. Figure 1-4 shows the results extrapolated to 400 °C (752 °F), which show the end-19 of-cycle RIP values for the IFBA rods were calculated to be higher than the values for 20 standard rods.

21 1-13 Figure 1-4 Calculated Rod Internal Pressures for the First 10 Cycles o f the Watts Bar 1 Nuclear Plant Unit 1 Reactor Under Vacuum Drying Conditions (P o = 133 Pa 2 (1 Torr); Tcladding(Max) = 400°C (752 °F) (Reproduction of Figure 4.24 from 3 Bratton et al. (2015)). The Data Points Shaded in Blue are for Standard Fuel 4 Rods (Approximately 21,000 Rods). The Data Points Shaded in Red are for 5 IFBA Fuel Rods (Approximately 47,000 Rods), with t he Lower Values for 6 IFBA Fuel with Annular Blanket Pellets and the Higher Values for Solid 7 Blanket Pellets 8 ORNL's calculated end-of-cycle RIPs from Figure 1

-4 were extrapolated to 25 °C (i.e. the 9 temperature for EPRI's EOL RIP data) to allow comparison with the results in Figure 1

-4. 10 Figure 1-5 shows t he aggregated RIP data at 25 °C (77 °F) within the burnup range of 40 to 11 74 GWd/MTU, along with best-fit and 3- Within the relevant burnup range of 40 12 to 62 GWd/MTU, the average +/- 3

-ues are 4.0

+/- 1.0 MPa ((5.8 +/- 1.4) x 10 2 psi) for standard 13 fuel rods with Zircaloy

-4 and ZIRLO cladding alloys. The 3

-(7.3 x 1 0 2 psi) is 14 a reasonable upper bound of data and calculated values for standard fuel rods. Furthermore , 15 reasonable upper bounds for data on IFBA rods are 7.9 MPa (1.1 x 10 3 psi) at 25 °C (77 °F) for 16 fuel designs with hollow (annular) pellets and 10.0 MPa (1.5 x 10 3 psi) for fuel designs with solid 17 pellets. 18 1-14 Figure 1-5 Aggregated Measured and Calculated Values for End-Of-Life Rod Internal 1 Pressures for PWR Fuel Rods.

Pressures are Presented at 25 °C (77 °F) 2 More recently, Richmond and Geelhood (2017) used FRAPCON to calculate EOL RIP for three 3 modern fuel designs with three representative dry storage thermal transients, each involving 4 drying operations with a peak cladding temperature of 400

°C (752 °F). The analyses 5 characterized the effects of fuel design to determine reasonably bounding cladding hoop 6 stresses.

The report provide s results for maximum EOL RIP for IFBA rods to be limited to 7 10.6 MPa (1.5 x 10 3 psi). These results suggest that, even at a peak cladding temperature of 8 400 °C (752 °F), the maximum cladding hoop stresses remain below 90 MPa (1.3 x 10 4 psi) for 9 the bounding ZIRLO-clad IFBA rods. At these pressures, the extent of hydride reorientation will 10 be very limited, if observed at all, which would indicate that the mechanical properties from the 11 as-irradiated condition will remain unchanged. The staff recognizes the discrepancies in the 12 results between Bratton et al

. (2015) and Richmond and Geelhood (2017) and is evaluating the 13 merits of both approaches used for the FRAPCON analyses. Until that evaluation is complete, 14 since the data of Bratton et al (2015) results in the highest EOL RIP and corresponding hoop 15 stresses, the staff will assume those values to be conservative and bounding when evaluating 16 the potential of hydride reorientation.

17 1.5.3.2 Gas Temperatures for Fuel Rods During Drying-Transfer , Storage And 18 Transportation 19 Peak temperatures for the gas inside fuel rods are highly dependent on the DSS or 20 transportation package design, fuel system design, fuel burnup, operating history, package 21 loading density, and the length of time the fuel was cooling in the spent fuel pool. Figure 1-6 22 0 1 2 3 4 5 6 7 8 9 10 11 40 45 50 55 60 65 70 75EOL PWR RIP at 25

°C (MPa)Rod Average Burnup (GWd/MTU)Data: Standard RodsFRAPCON: Standard RodsFRAPCON-IFBA-HBPFRAPCON-IFBA-SBPData Average + 3

-Data Average 1-15 shows the extrapolation of the upper-bound pressures shown in Figure 1-5 to rod internal gas 1 temperatures that may be experienced during drying

-transfer, dry storage and transportation.

2 The data are presented for the relevant burnup range of 45 to 62 GWd/MTU.

3 Figure 1-6 Calculated Rod Internal Pressure (Average +

3; 45 To 62 Gwd/MTU) a s a 4 Function o f the Spatially Averaged Gas Temperature for PWR Fuel Rods 5 (i.e., Standard Rods

), ZIRLO-Clad IFBA Rods with Hollow (Annular) Blanket 6 Pellets, a nd ZIRLO-Clad IFBA Rods with Solid Blanket Pellets 7 1.5.3.3 Peak Cladding H oop S tresses for Pressurized

-Water Reactor Fuel Rods During 8 Drying-Transfer a nd Storage/Transport Operations 9 ) is a function of the gas pressure difference across the cladding 10 wall (P i - P 0), where P i is the internal rod pressure and P o is the external pressure

, the cladding 11 inner diameter (D mi), and the cladding metal wall thickness (h m), as shown in Eqn.

1-1 for the 12 average hoop stress across the cladding wall (Figure 1-7). Gap closure is not considered.

13 = [D mi / (2

• h m)] (P i - P o) (Eqn. 1-1) 14 The geometrical parameter D mi/(2

• h m) will tend to increase with burnup due to waterside 15 corrosion of the cladding outer surface, which reduces h

m. Nominal as

-fabricated values of D mi 16 and h m for standard 17 x 17 PWR fuel rods are 8.36 mm and 0.57 mm, respectively, which 17 gives a geometrical factor of 7.33. However, fuel rod cladding is manufactured to specifications 18 with tolerances (i.e., +/- values) for the outer diameter and inner diameter and a minimum wall 19 thickness (e.g., >

0.56 mm). Thus, even for fresh fuel rods, the geometrical parameter will vary 20 somewhat along the length of one fuel rod and from fuel rod to fuel rod. For LBU SNF, the 21 decrease in h m is small and is partially counteracted by the decrease in D mi due to creep down.

22 4 6 8 10 12 14 16 18 20 22 25 50 75 100 125 150 175 200 225 250 275 300 325 350Pressure (MPa)Average Rod Internal Gas Temperature

(°C)Standard RodsIFBA (Hollow Blanket Pellets)IFBA (Solid Blanket Pellets) 1-16 For HBU SNF with creep out resulting from fuel-cladding mechanical interaction, D mi 1 approaches its as

-fabricated value and h m continues to decrease with burnup.

2 Figure 1-7 Fuel Cladding Tube with Stress Element Displaying Hoop Stress (), 3 Longitudinal Stress (z), Internal Pressure (P i), Cladding Thickness (H m), 4 External Pressure (P o), Circumferential Coordinate (, and Inner Cladding 5 Diameter (D mi) 6 Example calculations may be performed assuming that the EOL inner cladding diameter (D mi) is 7 8.36 mm (i.e., the as-fabricated value) with oxide layer thicknesses of 10 µm (e.g., M5), 40 µm 8 (e.g., ZIRLO), and 80 µm (e.g., Zircaloy-2, Zircaloy-4). The corresponding values of the 9 geometrical factor for these corrosion layers are: 7.41, 7.64, and 7.79.

These factors are used 10 in Eq n. 1-1 to generate the cladding hoop stress versus average gas temperature for several 11 values of corrosion layer thickness. The value of P o (varies from approximately 4 x 10-4 MPa 12 (3 Torr) for vacuum drying to less than 0.5 MPa (3.8 x 10 3 Torr) for storage) is assumed to be 13 zero for these calculations. Figure 1-8 shows that the hoop stress is a strong function of the 14 average gas temperature and a weaker function of the corrosion

-layer thickness. Calculated 15 cladding hoop stress values varied from 57 MPa (8.3 x 10 3 psi) at 180 °C (356 °F) for the 10

-µm 16 oxide case to 83 MPa (1.2 x 10 4 psi) at 340 °C (644 °F) for the 80

-µm oxide case.

17 1-17 Figure 1-8 Calculated Values of Cladding Hoop Stress (Average + 3; 45 To 62 1 Gwd/MTU) Vs. the Spatially

-Averaged Internal Gas Temperature for 2 Standard 17x17 PWR Fuel Rods With Corrosion Layers of 10 µm, 40 µm, 3 and 80 µm 4 For ZIRLO-clad IFBA rods, some 17 x 17 designs use smaller diameter cladding 5 (9.14 mm (0.360 in.)) and comparable cladding wall thickness (0.572 mm (0.023 in.)). This 6 design is called the "optimized fuel assembly." The geometrical factor for converting pressure to 7 hoop stress is 7.00 for as

-fabricated cladding and 7.28 for cladding with a 40

-µm corrosion 8 layer, which assumes the cladding inner diameter (8.00 mm (0.315 in.) remains the same as the 9 as-fabricated inner diameter for HBU SNF. Figure 1-9 shows that the cladding hoop stress for 10 the 40-µm corrosion

-layer case increases with gas temperature from about 90 MPa 11 (1.3 x 10 4 psi) to 120 MPa (1.7 x 10 4 psi) for IFBA fuel with hollow blanket pellets and from 12 about 110 MPa (1.6 x 10 4 psi) to 150 MPa (2.2 x 10 4 psi) for IFBA fuel with solid blanket pellets.

13 35 40 45 50 55 60 65 70 75 80 85 25 50 75 100 125 150 175 200 225 250 275 300 325 350PWR Rod Hoop Stress (MPa)Average Rod Internal Gas Temperature (°C) m oxidem oxidem oxide 1-18 Figure 1-9 PWR Hoop Stress a s a Function of Internal Gas Temperature for 17 x 17 1 IFBA Fuel Rods (for Both Hollow Blanket and Solid Blanket Pellets) with a 2 40-µm Corrosion Layer 3 The above discussion provide s a technical basis for the staff to defin e conservative bounding 4 cladding hoop stress conditions for the testing of HBU SNF mechanical performance, as the 5 assessment in Chapter 2 will discuss

. 6 1.5.4 Ring Compression Testing 7 Ring compession testi ng (RCT) has been conducted in the United States and Japan to assess 8 residual ductility of cladding with reoriented hydrides following pinch loads (Aomi et al

., 2008; 9 Billone et al., 2013; Billone et al., 2014; Billone et al., 2015). RCT of zirconium

-based cladding 10 alloys has shown reduced ductility when subjected to pinch loads at a sufficiently low 11 temperature; this temperature has been generally referred to as a ductile

-to-brittle transition 12 temperature or ductility transition temperature (DTT)

. 13 In previous NRC-sponsored research, Argonne National Laboratory (ANL) sectioned rings from 14 pressurized

-and-sealed rodlets fabricated with cladding from ZIRLO

-clad and Zircaloy clad 15 fuel rods irradiated to high burnup (beyond the NRC's peak rod licensing limit i n 16 commercial PWRs) (Billone et al., 2013)

(Figure 1-1 0). These rodlets had been heated to a 17 peak temperature of 400 °C (752 °F) (consistent with the guidance limit in ISG

-11, Revision 3 18 (NRC, 2003 a) and held at this temperature for 1 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> with variable target hoop stresses 19 (110 MPa (1.6 x 10 4 psi), 140 MPa (2.0 x 10 4 psi)), and then slow-cooled at 5

°C/h (9 °F/h) 20 under conditions of decreasing pressure and hoop stress. Metallographic examination of one 21 cladding ring surface per rodlet was used to quantify the degree of radial hydride precipitation in 22 terms of the average length of radial hydrides. Several other rings were used to determine the 23 40 60 80 100 120 140 160 25 50 75 100 125 150 175 200 225 250 275 300 325 350PWR Rod Hoop Stress (MPa)Average Rod Internal Gas Temperature (°C) IFBA (Hollow Blanket Pellets)IFBA (Solid Blanket Pellets) 1-19 average hydrogen content of the rodlet, along with circumferential and axial variations in 1 hydrogen conten t. Up to four rings were subjected to RCT to induce pinch loads at test 2 temperatures from 20

°C (68 °F) to 200 °C (392 °F). 3 Figure 1-10 RCT of a Sectioned Cladding Ring Specimen in ANL's Instron's 8511 Test 4 Setup. Tests Were Conducted in the Displacement

-Controlled Mode to a 5 1.7-Mm Maximum Displacement in a Controlled Temperature Environment (6 p = RCT Offset Displacement at 12 O'clock Position Relative to Static 7 Support at 6 O'clock

D mo = Outer Diameter of Cladding Metal

p/D mo = 8 RCT Offset Strain (Percent)) (Reproduction Of Figure 6 From Billone et al., 9 2012)) 10 RCT load-displacement curves were used to determine the offset displacement (normalized to 11 the pretest sample outer diameter to give offset strain) as a function of test temperature. The 12 offset strain was plotted against test temperature for each rodlet to determine the DTT 13 (see Figure 1-1 1). Post-RCT metallographic examinations were also performed to determine 14 the number and extent of cracks that had formed, as well as to generate additional data for the 15 degree of radial hydride precipitation (Billone, et al., 2013).

16 To define ductility per RCT load-displacement data, a 2

-percent offset strain (p/D mo) before a 17 crack extend ed through more than 50 percent of the cladding wall thickness was chosen as the 18 figure of merit for the transition between ductile and brittle behavior (Billone et al., 2013). Figure 19 1-1 1 shows representative deformation (i.e., offset strain) curves as a function of the alloy, peak 20 hoop stress at a 400 °C (752 °F) peak cladding temperature, and actual RCT temperature. The 21 figure also shows the radial hydride continuity factor (RHCF), which represents the effective 22 radial length of continuous radial

-circumferential hydrides normalized to the wall thickness. ANL 23 used the RHCF as a figure of merit f o r determining the degree and severity of radial hydrid e 24 precipitation. The radial hydrides in Zircaloy

-4 HBU SNF ring specimens were relatively short 25 (i.e., RHCF of 9 percent for a peak hoop stress of 110 MPa (1.6 x 10 4 psi), and 16 percent for a 26 peak hoop stress of 140 MPa (2.0 x 10 4 psi)) and the ductility increased gradually with 27 temperature. In ZIRLO-clad HBU SNF ring specimens

, the radial hydrides were longer (i.e., 28 RHCF of 30 percent for a peak hoop stress of 110 MPa (1.6 x 10 4 psi), and 65 percent for a 29 1-20 peak hoop stress of 140 MPa (2.0 x 10 4 psi)) and the ductility increased sharply with the 1 increase in RCT temperature. ANL fit th e limited ZIRLO data points with S

-shaped curves 2 (hyperbolic tangent functions) typical of materials that exhibit a ductile

-to-brittle transition. The 3 data show that the DTT shifted from around room temperature in a cladding material with short 4 radial hydrides to higher values in a cladding material with longer radial hydrides. The limited 5 data also indicates a trend of lower DTTs for materials with lower peak cladding stresses. 6 Figure 1-11 Ductility vs. RCT for Two PWR Cladding Alloys Following Slow Cooling 7 from 400°C (752 °F) at Peak Target Hoop Stresses of 110 Mpa (1.6 x 10 4 Psi) 8 and 140 Mpa (2.0 x 10 4 Psi) (From Billone e al., 2013) 9 ANL also conducted RCT research under DOE sponsorship.

It obtained results for the following 10 (Billone et al., 2014; Billone et al., 2015)

11 HBU Zircaloy-4 in the as

-irradiated condition with moderate

-to-high hydrogen content 12 HBU ZIRLO in the as

-irradiated condition and following simulated drying

-storage at peak 13 temperatures of 400

°C (752 °F) and 350 °C (662 °F) with peak hoop stresses from 14 80 MPa (1.2 x 10 4 psi) to 94 MPa (1.4 x 10 4 psi) 15 HBU M5 in the as

-irradiated condition and following simulated drying

-storage at 16 400 °C (752 °F) with peak hoop stresses of 90 M Pa (1.3 x 10 4 psi), 110 MPa 17 (1.6 x 10 4 psi), and 140 MPa (2.0 x 10 4 psi) 18 ANL conducted two additional tests with HBU ZIRLO cladding subjected to three drying cycles 19 (e.g., from 400 °C (752 °F) to 300 °C (572 °F) and from 350

°C (662 °F) to 250 °C (482 °F)) at 20 peak hoop stress of about 90 MPa (1.3 x 10 4 psi). The latter results suggest that multiple drying 21 0 2 4 6 8 10 12 14 0 25 50 75 100 125 150 175 200 225Offset Strain (percent)RCT Temperature (°C)9+/-5% RHCF; Zry-4 @ 110 MPa16+/-4% RHCF; Zry-4 @140 MPa 30+/-12% RHCF; ZIRLOŽ @ 110 MPa 65+/-17% RHCF; ZIRLOŽ @140 MPa Brittle Ductile 1-21 cycles ha ve no effect on the length of radial hydrides or the DTT at this low stress level.

Figur e s 1 1-1 2 through 1-1 4 show results for Zircaloy

-4 , ZIRLO , and M5 in both as

-irradiated and hydride

-2 reoriented condition following cooling from 400°C (752 °F) (Billone et al., 2014; Billone et al., 3 2015). 4 Figure 1-12 Ductility Data, as Measured by RCT , for A s-Irradiated Zircaloy-4 and 5 Zircaloy-4 Following Cooling from 400 °C (752 °F) Under Decreasing 6 Internal Pressure and Hoop Stress Conditions (From Billone e t al., 2013) 7 0 2 4 6 8 10 12 14 0 25 50 75 100 125 150 175 200 225Offset Strain (percent)RCT Temperature (°C)300+/-15 wppm H640+/-140 wppm H520+/-90 wppm H615+/-82 wppm HBrittleDuctile145 MPaat 400°C113 MPaat 400°CHigh Burnup Zircaloy-4 As-Irradiated 1-22 Figure 1-13 Ductility Data, as Measured by RCT , for a s-Irradiated ZIRLO a nd ZIRLO 1 Following Cooling from 400 °C (752 °F) Under Decreasing Internal Pressure 2 and Hoop Stress Conditions (From Billone e t al., 2013) 3 0 2 4 6 8 10 12 14 0 20 40 60 80 100 120 140 160 180Offset Strain (percent)RCT Temperature (°C)530+/-70 wppm H535+/-50 wppm H530+/-115 wppm H480+/-131 wppm H385+/-80 wppm HBrittleDuctile111 MPaat 400°C80 MPaat 400°CHigh Burnup ZIRLO As-Irradiated89 MPaat 400°C 1-23 Figure 1-14 Ductility Data , as Measured by RCT , for As-Irradiated M5 a nd M5 Following 1 Cooling from 400 °C (752 °F) Under Decreasing Internal Pressure and Hoop 2 Stress Conditions (From Billone et al., 2013

) 3 The staff recognizes the uncertainties associated with the ductility curve fits of ANL's RCT data 4 because of the limited number of data points. However, the limited results appear ed to support 5 the following general conclusions: (1) the DTT generally increases with increasing hoop 6 stresses (i.e., the ductility transition shifts to higher cladding temperature), (2) both the 7 susceptibility to radial hydride precipitation and ductility changes depend on cladding type and 8 initial hydrogen content, and (3) depending on the cladding and test conditions, the DTT can 9 occur at temperatures in the range of 20 °C (68 °F) to 185 °C (365 °F). The results for as

-10 irradiated Zircaloy

-4 are consistent with studies by Wisner and Adamson (1998) and Bai et al 11 (1994). The staff considered these conclusions when defining limiting conditions for inducin g 12 radial hydrides and conducting fatigue and bending testing of HBU SNF (see Chapter 2).

13 It is important to note that th e DTT is not an intrinsic property of a cladding alloy material with a 14 given homogeneous composition, in the classical metallurgical sense, but it is highly dependent 15 on the composite microstructure (hydride

-zirconium matrix, as determined by reactor operating 16 conditions

), fabrication conditions (degree of cold working, recrystallization) and the operating 17 conditions during drying

-transfer, storage or transportation (peak cladding temperature, peak 18 hoop stress, temporal cooling profile

). Further, the DTT was established based on an arbitrarily

-19 defined performance criterion (e.g., 50 percent cladding through

-wall crack prior to 2

-percent 20 offset strain deformation), and based on a limited number of data points for each cladding alloy.

21 It is also important to note that, due to the radial and axial temperature gradients in a DSS or 22 transportation package, it is highly likely that only a small fraction of the cladding in a given 23 assembly will reach high enough temperatures and hoop stress es to have sufficient hydride 24 reorientation during cooling. Those hotter axial locations of the cladding will likely be the last to 25 reach a DTT during transpor

t. 26 0 2 4 6 8 10 12 14 0 25 50 75 100 125 150 175 200 225Offset Strain (%)RCT Temperature (°C)76+/-5 wppm H58+/-15 wppm H (90 MPa)72+/-10 wppm H (111 MPa)94+/-4 wppm H (142 MPa)142 MPa at 400°C111 MPaat 400°CDuctileBrittleHigh Burnup M5 As-Irradiated90 MPaat 400°C 1-24 1.5.5 Staff's Assessment of Ring Compression Testing Results 1 As previously discussed, the staff has long expected that hydride reorientation would not 2 compromise cladding integrity due to fuel rod bending (i.e., bending expected during normal 3 conditions of storage and transport

), since the principal tensile stress field associated with rod 4 bending caused by lateral inertia loads is parallel to both radial and circumferential hydrides 5 (Tang et al

., 2015). The staff has considered that any reduced cladding ductility due to hydride 6 reorientation could only potentially compromise the analyzed fuel configuration for pinch loads 7 experienced during drop accident scenarios, if the fuel ha d significantly cooled during the 8 transportation period. More specifically, the staff had expressed concern that reorientation 9 could decrease failure stresses and strains in response to transportation

-induced pinch loads 10 during a 9-m (30-ft) drop scenario as a result of rod-to-grid spacer contact, rod-to-rod contact, or 11 rod-to-basket contact. 12 To address the concern of reduced ductility during drop accidents, the staff previously proposed 13 varied approaches to demonstrate that the failure limits for as

-irradiated cladding as used in the 14 design-bas is structural evaluations would continue to be adequate even if hydride reorientation 15 occurred. One of these approaches was based on justifying an RCT-measured DTT for each 16 cladding alloy in the proposed fuel contents, and demonstrating that the minimum cladding 17 temperature remained above the RC T-measured DTT for the entire duration of transport. The 18 minimum cladding temperature assumed for transport operations would need to be bounding to 19 the contents upon consideration of the cold temperature requirement in 10 CFR 71.71(c)(2), i.e.

20 an ambient temperature of

-40 °C (-40 °F) in still air and shade. If these conditions were met, 21 then mechanical properties of the as

-irradiated cladding material (i.e., material that did not 22 account for the precipitation of radial hydrides

), would be considered adequate for the structural 23 evaluation

. 24 As an alternative approach, if the applica nt could not reasonably demonstrate that sections of 25 the fuel cladding remained above the RCT-measured DTT during the entire duration of 26 transport, the staff propose d that the application provide additional safety analyses assuming 27 hypothetical reconfiguration of the HBU fuel contents. If neither of these two approaches i s 28 satisfactory for demonstrating compliance with 10 CFR Part 71 regulations,then the staff would 29 expect that t he fuel would be canned an d classified as damaged

. 30 Since proposing these approaches, the staff has reevaluated whether results from RCT of 31 defueled specimens are accurately representative or if they are overly conservative relative to 32 the actual hoop

-loading conditions experienced by the fuel during a 9-m (30-ft) drop. During 33 RCT, the circumferential (hoop) tensile bending stress is perpendicular to the plane of the radial 34 hydrides, which is different from the relative orientation of the applied stress and hydrides under 35 axial tensile bending where the longitudinal (axial) tensile bending stress is always parallel to 36 the plane of both the circumferential and radial hydrides. The orientation of the tensile stress is 37 expected to make a difference in the response of the cladding.

38 The RCT defined a D TT used to determine cladding failure due to pinch loads. However, it is 39 necessary to consider the importance of this failure mode in the determination of cladding 40 integrity in the event of a drop accident. To do this, the RCT must be examined for what it is, a 41 test in which diametrically

-opposed concentrated compressive forces are applied to a fuel 42 cladding longitudinal segment that does not contain fuel. During NCT and HAC side drops, the 43 fuel rod is loaded by lateral inertia loads that are resisted by distributed loads applied to the 44 bottom of the rod at the flexible grid spacer springs (Figure 1-1 5). Further, the inertia load in the 45 1-25 rod is transferred to the grid spacer support as a shear force in the cladding (and pellets) not as 1 a concentrated load at the top of the rod.

2 Single Rod Model Single Assembly Model Figure 1-15 Geometric Models for Spent Fuel Assemblies i n Transportation Packages 3 (Reproduction , i n Part, Of Figure 10 from Sanders e t al., 1992) 4 Given that the forces and displacements in the RCT are measurably different from the actual 5 forces and displacements applied to the rod at the grid spacer support, it is not likely that the 6 pinch-mode of failure will play a significant role in undermining cladding integrity. To quantify 7 the difference between these loading cases, the staff analyzed two ring segments for different 8 loading conditions and the change in diameter calculated. In the first case the ring segment 9 was loaded by diametrically

-opposed compressive forces like those of RCT (Case 1, Table 17, 10 Roark and Young (1975)). In the second case the ring segment was supported at the bottom by 11 a concentrated reaction and loaded by a downward load uniformly distributed around the 12 circumference of the ring to simulate a shear loading as in a side drop (Case 13, Table 17, 13 Roark and Young (1975)). In both cases the total applied load was the same.The ratio of the 14 change in diameter of the second case to the first case is 0.48. Thus, the diametrically

-opposed 15 compressive forces produced more than twice the displacement when compared to the 16 circumferentially distributed load. In addition, at the pellet

-cladding interface

, the pellet and 17 cladding are bonded and, thus a gap cannot exist between the

m. Thus , the staff considers that, 18 under a pinch load, ovalization of the cladding cross

-section is very unlikely and any 19 circumferential bending stress that does exist will be negligible. The RCT conducted to dat e 20 does not account for the rod's resistance to ovalization provided by the pellet.

21 Based on the RCT load-displacement data, ANL defined a figure of merit for cladding ductility 22 (i.e., the transition between ductile and brittle behavior) to be a 2-percent offset strain prior to 23 a crack extending through more than 50 percent of the cladding wall (Billone et al., 2013). If 24 the strains experienced during RCT's diametrically

-opposed loads result in twice those that 25 would be experienced during lateral inertial loads, then the DTT is likely to shift to lower 26 temperatures (potentially room temperature or lower). Therefore, the staff considers that the 27 DTT defined by RCT experiments is overly conservative and not representative of actual fuel 28 and stress conditions during NCT and HAC drop scenarios. The DOE is planning on 29 sponsoring a research program in which 25 HBU fuel rods will undergo testing to determine 30 their characteristics, material properties, and rod performance following representative drying

-31 transfer and cooldown (Hanson et al

., 2016). The staff expects that material property testing 32 conducted under this program will provide confirmation that the cladding displacements 33 experienced by fueled cladding specimens during RCT will be lower than those measured in 34 1-26 defueled specimens and that ductility during accident drop scenarios is not compromised.

1 Results from the static and fatigue bend testing discussed in Chapter 2 further justify the 2 staff's conclusion that the pellet imparts structural support to the mechanical performance of 3 the fuel rod.

4 2-1 2 ASSESSMENT OF STATIC BENDING AND FATIGUE STRENGTH 1 RESULTS ON HIGH BURNUP SPENT NUCLEAR FUEL 2 2.1 Introduction 3 The sealed canister or cask cavity serves as the primary barrier in a dry storage system (DSS) 4 or transportation package for protecting against the release of radioactive solid particles or 5 gases from the loaded spent nuclear fuel (SNF) to the atmosphere. The spent fuel cladding 6 also serves as a confinement or containment barrier for preventing radioactive solid particles 7 and fission gasses from being released into the interior cavity of the DSS or transportation 8 package. The cladding not only provides a barrier for preventing the release of radioactive 9 material but also prevents fuel reconfiguration during storage and transport operations.

10 Therefore, the integrity of the cladding is an essential component of a defense

-in-depth strategy 11 to protect the public health and safety.

12 Until recently, research to understand the structural behavior of spent fuel rods during 13 transportation and storage has focused entirely on obtaining mechanical and strength properties 14 of spent fuel cladding. As a result, the flexural rigidity and structural response of fuel rods 15 during normal and accident events ha ve been based on the mechanical and strength properties 16 of only the cladding. The contribution of the fuel pellets to increasing the flexural rigidity of the 17 rod has been neglected. However, recent research discussed in NUREG/CR

-7198 , Revision 1 , 18 "Mechanical Fatigue Testing of Hig h-Burnup Fuel for Transportation Application," issued 19 October 2017 (NRC, 2017a), on the static bending response and fatigue strength of fuel rods 20 considered as a composite system of cladding and fuel pellets, has begun to provide some of 21 the necessary data to allow a more accurate assessment of the structural behavior of the 22 composite fuel rod system under normal conditions of transport (NCT) and hypothetical accident 23 conditions (HAC), as well as DSS drop and tip

-over events.

24 2.2 Cyclic Integrated Reversible Fatigue Tester 25 In 2009, the U.S. Nuclear Regulatory Commission (NRC) tasked Oak Ridge National Laboratory 26 (ORNL) with investigating the flexural rigidity and fatigue life of high burnup (HBU) SNF 27 (N RC, 2017a). The testing was designed to evaluate the fuel rod as a composite system, 28 including the presence of intact fuel inside the cladding and any pellet/cladding bonding effects.

29 The project proceeded in two phases. Phase I involved testing HBU SNF in the as-irradiated 30 state, where hydrides are expected to be predominantly in the circumferential

-axial orientation.

31 Phase II involved testing HBU SNF segments subjected to a treatment designed to reorient the 32 hydrides in the cladding to be predominantly in the radial

-axial orientation. All testing was 33 conducted at room temperature, which is expected to result in the most limiting cladding 34 ductility.

35 In response to the NRC tasking, in 2011, ORNL proposed a bending fatigue system for testing 36 HBU SNF rods. The system is composed of a U

-frame equipped with load cells for imposing 37 pure bending loads on the SNF rod test specimen and measuring the in-situ curvature of the 38 fuel rod during bending using a set

-up of three linear variable differential transformers (LVDT) 39 (Figure 2-1). Pure bending is a condition of stress in which a bending moment is applied to a 40 beam without the simultaneous presence of axial, shear, or torsional forces.

41 2-2 Figure 2-1 Horizontal Layout of ORNL U-Frame Setup (Top), Rod Specimen and Three 1 Lvdts for Curvature Measurement (Mid dle), and Front View of CIRFT 2 Installed in ORNL Hot Cell (Bottom) (Figure 4 from NUREG/CR-7198, 3 Revision 1 (NRC, 2017a

)) 4 SNF rod End-blocks LVDT clamp Three LVDTs for curvature measurement Rod specimen Rigid arms Connecting plates Universal testing machine links Load cell (middle) (bottom) (top) 2-3 On August 19, 2013, a testing system was installed in a hot cell at ORNL's Irradiated Fuels 1 Examination Laboratory and formally named the "cyclic integrated reversible

-bending fatigue 2 tester" (CIRFT). After tuning of the test system and performance of benchmark testing in 3 September 2013, Phase I testing began on HBU SNF rod segments with intact Zircaloy-4 4 cladding irradiated in the H.B. Robinson Steam Electric Plant (HBR). ORNL completed four 5 static tests under displacement control at the rate of 0.1 mm/s to a maximum displacement of 6 12.0 mm. In early November 2013, the benchmark and static test results were critically 7 reviewed at a meeting between representatives from the NRC and ORNL. Dynamic testing was 8 then initiated, and 16 cyclic tests were completed in the Irradiated Fuels Examination 9 Laboratory

. Load ranges applied to the CIRFT varied , to produce bending moments in the rod, 10 from +/-5.08 to +/-35.56 N*m. There were 12 dynamic tests with rod fracture and 4 tests without 11 rod fracture

. One of the cyclic tests reached 1.3 x 10 7 cycles with no rod fracture. The test was 12 terminated as higher cycles would not be expected during actual transport.

13 Phase II testing began in 2016, again using HBR HBU SNF rods with intact Zircaloy

-4 cladding, 14 which had been subjected to an aggressive hydride reorientation treatment (HRT)

(see 15 Section 2.3.4). ORNL completed testing on four specimens in the CIRFT following an HRT: 16 one in static loading (hereafter referred to as HR2)

, and three in dynamic loading (hereafter 17 referred to as HR1, HR3 , and HR4). The fatigue lifetime and flexural rigidity of these samples 18 were compared to the results obtained in Phase I for as-irradiated samples.

19 The following observations can be made about the results of the static testing:

20 The HBR HBU SNF rods in the as

-irradiated state exhibited a multiple

-stage constitutive 21 response, with the two linear stages followed by a nonlinear stage. The flexural rigidity at 22 2, corresponding to an elastic modulus of 101 to 23 125 GPa. The flexural rigidity at the second stage was 55 to 61 N*m 2, and the 24 corresponding elastic modulus was 88 to 97 GPa.

25 Most HBR HBU SNF rods in the as

-irradiated state under static unidirectional loading 26 fractured at a location coincident with the pellet

-to-pellet interface, as validated by the 27 posttest examinations showing pellet end faces in most of the fracture surfaces.

28 Fragmentation of the pellets also occurred to a limited degree, along with cladding 29 failure. 30 The static CIRFT results indicate a significant increase in a fueled SNF rod's flexural 31 rigidity compared to a calculated response for cladding only. This applied to both as

-32 irradiated and HRT SNF rods.

33 For the HBR HBU SNF rods, the static CIRFT test results show that at bending moments 34 less than 30 N*m the flexural rigidities of the as

-irradiated rods and the HRT HR2 rod are 35 essentially the same.

36 The sample subjected to a n HRT and tested under a static bending load showed 37 reduced flexural rigidity at higher loads compared to as

-irradiated samples.

38 Nevertheless, material tested in the as

-irradiated and HRT state both had higher flexural 39 rigidity than the calculated cladding

-only response.

40 The static CIRFT test result for HR2 supports the pretest expectation (hypothesis) that 41 because the tensile bending stress in the cladding is parallel to the plane of both the 42 radial and circumferential hydrides, the presen ce of radial hydrides would not 43 2-4 significantly alter the flexural response when compared to the case where only 1 circumferential hydrides are present.

2 The methodology developed in this study calculate cladding stress and strain is 3 applicable to all cladding types, and the use of cladding

-only properties to calculate 4 cladding stress and strain is always conservative for all cladding types.

5 The HBR HBU SNF rods in the as

-irradiated state survived static unidirectional bending 6 to a maximum curvature of 2.2 to 2.5 m

-1, or a maximum m7 maximum static unidirectional bending values were bounded by the CIRFT device 8 displacement capacity. The maximum equivalent strain was 1.2 to 1.4 percent. 9 Based on the static CIRFT test results, the lower

-bound safety margin against fuel rod 10 failure during a n HAC side drop event is 2.35 assuming the side drop imparts a 50

-g 11 load to the package body (see Section 2.3.4.2). 12 The following observations can be made about the results of the dynamic testing:

13 The fatigue life of HBR HBU SNF rods in the as

-irradiated state in the cyclic tests 14 depended on the level of loading. Under loading with moments of +/-8.89 to +/-35.56 15 -namely +/-0.03 to +/-0.38 percent strain- the fatigue life ranged from 5.5 x 10 3 to 16 2.3 x 10 6 cycles. 17 The -N curve of the HBR HBU SNF rods in the as

-irradiated state can be described by 18 a power function of y = 3.839*x-0.298, where x is the number of cycles to failure, and y is 19 the strain amplitude (percent). 20 Maxima of the curvature during dynamic tests in the as

-irradiated state ranged from 21 +/-0.08 to +/-0.78 m

-1. The -N curve of the HBR HBU SNF rods can be described by a 22 power function of y = 6.864*x-0.283, where x is the number of cycles to failure, and y is the 23 -1). A fatigue limit is likely located 24 between 0.08 and 0.1 3 m-1 if it is defined at 10 7 cycles (as is typical for material fatigue 25 endurance limits).

26 The failure of HBR HBU SNF rods under cyclic loading often occurred near pellet

-to-27 pellet interfaces.

28 The samples subjected to a n HRT showed a slightly reduced lifetime compared to as

-29 irradiated samples in dynamic testing (see Section 2.3.3

). 30 The following sections provide an assessment by the NRC staff (the staff) of ORNL's CIRFT 31 data and present conclusions as to the expected structural performance of HBU SNF during dry 32 storage and transportation

. 33 2.3 Application of the Static Test Results 34 When evaluating the HAC 9-m (30-ft) drop test, as required by Title 10 of the Code of Federal 35 Regu lations (10 CFR) 71.73(c)(1), two drop orientations produce distinctly different structural 36 behaviors in the fuel rods. These orientations are the side drop and the end drop (Figure 2-2). 37 In the side drop, lateral inertia loads are applied to the fuel rods

, and bending dominates the 38 structural response. In the end drop, axial compression and the associated buckling of the fuel 39 2-5 rod dominates the structural response. For a side-drop event, the CIRFT static bending test 1 results from NUREG/CR

-7198, Revision 1 (NRC, 2017a

), can be directly applied to quantify the 2 fuel rod structural response. For the end drop, the presence of axial compression in the fuel rod 3 represents a force component that was not present in the CIRFT static bending tests. This, 4 however, does not pose a problem since the CIRFT static test results can be used to 5 conservatively quantify the effect of the fuel pellets on increasing the flexural rigidity of the rods 6 to resist buckling. 7 Figure 2-2 Schematic Diagram of End and Side Drop Accident Scenarios 8 (Revised Figure 5-168 from Patterson and Garzarolli (2015)) 9 2.3.1 Spent Fuel R od Behavior i n Bending 10 The behavior of a fuel rod in bending generally depends on three things:

(1) the type of loading

, 11 (2) the bond between the cladding and fuel

, and (3) the behavior of the pellet

-pellet interface.

12 Fundamentally

, there are two types of bending

-bending without shear and bending with shear.

13 Bending without shear is pure bending (i.e., constant moment or curvature, as exhibited in the 14 ORNL CIRFT tests) and produces no shear stress at the interface between the cladding and 15 fuel pellet. Pure bending is a special case that does not often occur in practice. What occurs 16 more often is the case of a laterally

-supported fuel rod subjected to a transverse inertia loading, 17 as in a side drop, where the rod is subjected to both bending and shear forces.

1 Although both 18 bending and shear are acting, the structural response would be expected to be different

, 19 depending on whether the cladding is bonded to the fuel pellet.

20 1 Because the fuel behaves in a brittle manner while the cladding behaves in a ductile manner, all of the bending tensile stresses will occur in the cladding.

The cladding and fuel will resist the shear forces, but for simplicity

, it can be conservatively assumed that all of the shear is resisted by the cladding. A simple calculation shows that during a side drop event

, the uniformly loaded fuel rod spanning over multiple grid spacers will have maximum tensile stresses due to bending that are more than an order of magnitude greater than the maximum tensile stresses due to shear. Therefore, bending dominates the response of the fuel rod, and this is why the CIRFT tests can accurately represent the behavior of an actual fuel rod during a side drop event.

2-6 2.3.2 Composite Behavior of a Spent Fuel R od 1 The normal explanation for the structural response of the fuel ed-rod composite system is as 2 follows. If the pellet is not bonded to the cladding, displacement compatibility is not maintained 3 at the pellet

-cladding interface

, and composite action does not occur. In this case

, the flexural 4 rigidity is given by the following

5 EI = E c I c + E p I p (E q n. 2-1) 6 That is, the flexural rigidity is equal to the sum of the individual flexural rigidities of the cladding 7 and fuel pellets, where E c and I c are the elastic modulus and moment of inertia of the cladding, 8 respectively, and E p and I p are the elastic modulus and moment of inertia of the pellet, 9 respectively. On the other hand, if the pellet is bonded to the cladding, displacement 10 compatibility is maintained at the pellet

-cladding interface and composite action occurs. In this 11 case , the flexural rigidity is calculated by transforming the pellet properties into equivalent 12 cladding properties (i.e., by multiplying the pellet moment of inertia by E p/E c). This is the same 13 technique commonly used for reinforced concrete (Winter and Nelson, 1979). As mentioned 14 above, a spent fuel rod is a composite system consisting of cladding and spent fuel. To fully 15 understand the unique behavior of this composite system, the bending behavior of a more 16 general composite beam will be discussed. Consider a composite concrete and steel I

-beam 17 where a concrete slab, rectangular in cross

-section, is poured onto the top flange of a steel I

-18 beam (Figure 2

-3). This type of composite beam is commonly found in highway bridge 19 construction. Assume the concrete and steel beam are simply supported and a concentrated 20 load is applied at mid

-span. If the concrete slab and steel beam are not bonded to each other, 21 no shear transfer takes place at the interface between the steel and concrete, and the flexural 22 rigidity (EI) is equal to the sum of the individual flexural rigidities of the concrete slab and steel 23 beam taken separately.

24 Figure 2-3 Typical Composite Construction of a Bridge 25 On the other hand, if the concrete slab and steel beam are bonded to each other, as typically 26 done using shear studs, then shear transfer takes place and the concrete slab and steel beam 27 act as a composite section. In this case

, the flexural rigidity of the composite beam will be 28 significantly greater than the sum of the individually flexural rigidities taken separately. This 29 2-7example of a concrete slab bonded to the top flange of a steel beam illustrates the behavior of a 1 composite system where the centers of gravity of each of the two components (i.e., concrete 2 slab and steel I

-beam) are not coincident.

3 For the special case where the centers of gravity of the two components are coincident, the 4 flexural rigidity of the composite section is always equal to the sum of the flexural rigidities of the 5 individual components regardless of whether the components are bonded or unbonded. The 6 following example illustrates this concept. Consider a simply supported span co mp osed of two 7 beams, each with a rectangular cross

-section 2 in. wide, and 6 in. deep (i.e., a "2 x 6"). Let the 8 2 x 6's be configured one on top of the other, where the centers of gravity (cg s) are not 9 coincident as shown in Figure 2-4a. If the beams are unbonded, the moment of inertia of the 10 section (I = bh 3/12 per beam), is equal to: 2 x 2 in. x (6 in.)3/12 = 72 in

.4. If they are bonded, 11 then the moment of inertia of the section is equal to: 2 in. x (2 x 6 in.)3/12 = 288 in.4 12 Figure 2-4 Influence of cg Position on Composite Beam Stiffness:

13 (a)cgs Are Not Coincident, (b) cgs Are Coincident 14 Now let the 2 x 6s be configured as shown in Figure 2-4b, where the cgs are aligned on the 15 same bending axis (i.e., they are "coincident

"). If they are unbonded, the moment of inertia of 16 the section is: 2 x 2 in. x (6 in.)3/12 = 72 in.4. If they are bonded I = 2 x 2 in. x (6 in.)3/12 = 72 17 in.4 Thus, w hen the cgs of the 2 x 6's are "coincident" the flexural rigidity of t he beam i s the 18 sum of the individual flexural rigidities of the 2 x 6's regardless of whether the 2 x 6s are bonded 19 or unbonded. While previously unrecognized, this is the situation with a spent fuel rod, where 20 the cladding cylindrical tube and the spent fuel cylindrical solid section have coincident c gs. 21 Thus, for a spent fuel rod, where the fuel is a homogenous solid, the flexural rigidity is given by 22 Equation 2-1, regardless of whether the fuel is bonded to the cladding. All moments of inertia 23 are taken about the neutral axis of the fuel rod.

24 Calculation of Cladding Strain 25 The objective of this section is to develop a simple methodology that uses the CIRFT static test 26 data for fully-fueled composite spent fuel rods to evaluate spent fuel rod cladding strain. T he 27 methodology presented here to determine cladding response (i.e., cladding stresses and 28 strains) is based on a set of assumptions that are consistent with those made by ORNL in its 29 2-8 presentation of CIRFT results in NUREG/CR-7198, Revision 1 (NRC, 2017a). These 1 assumptions, which are discussed in greater detail below, are based on the integrated average 2 response of the fuel rod along its gauge length.

3 Figure 2-5 Images of Cladding

-Pellet Structure in HBU SNF Rod (66.5 Gwd/MTU, 40-4 70 µm Oxide Layer, 500 Wppm H Content i n Zircaloy-4): (a) Overall Axial 5 Cross Section a nd (b) Enlarged Area (Revised Figure 33 from NUREG/CR-6 7198, Revision 1 (NRC, 2017a)) 7 The fuel rod composite system (Figure 2-5) is composed of cladding, which exhibits ductile 8 behavior, and the fuel pellet, which exhibits brittle behavior. In a spent fuel rod subject to 9 bending, where the fuel is a homogenous solid, the neutral axis is at the center of the rod cross

-10 section , provided that the brittle fuel does not crack in tension. Once the fuel cracks, the neutral 11 axis will shift toward the compression side of the cross

-section. The ORNL tests show that the 12 region of the fuel weakest in tension is at the pellet

-pellet interface. When the pellet

-pellet 13 interface cracks, the tensile stress in the cladding at the crack face

, will increase significantly.

14 On either side of the crack face the shear stress between the cladding and fuel is high an d 15 decreases parabolically with distance from the crack (Figure 2-6). The high tensile stress in the 16 cladding at the crack face also decreases parabolically with distance from the crack. Thus

, the 17 cladding tensile stresses will vary significantly along the length of the rod

they are highest at the 18 crack face and much lower away from the crack face. Even though this behavior is known to 19 occur, only the average tensile bending stress can be calculated from the static test results 20 because the measured curvature is the integrated average curvature over the measurement 21 length (gauge length) of the rod.

22 2-9 Figure 2-6 Approximate Extreme Fiber Tensile Stresses Between Pellet

-Pellet Crack 1 The LVDTs measure displacements at three locations on the test specimen. The distance 2 between the first and third probes is the gauge length of the specimen. Because the bending 3 moment is constant along the gauge length, it would be expected that several pellet

-pellet 4 interface cracks would develop within the gauge length. That being the case, the cladding 5 tensile stresses and strains along the gauge length will vary significantly. However, this 6 variation in strain along the gauge length was not, and cannot be, measured. What was 7 measured is the average curvature along the gauge length. Therefore, only the average tensile 8 strain (i.e., the smeared tensile strain) can be calculated. The average tensile strain , , along 9 the gauge length is equal to the curvature , , multiplied by the distance to the neutral axis, ymax: 10 =

• ymax (Eq. 2-2) 11 However, ymax can vary significantly along the gauge length. At a section where the fuel has not 12 cracked, ymax is equal to the outer radius, r. At a pellet

-pellet interface crack, ymax would be 13 greater than the radius but less than the diameter. However, because the measured and 14 calculated results are averages over the gauge length, a convention must be adopted for 15 calculating cladding strain and this convention must be consistently applied throughout. The 16 convention used in NUREG/CR

-7198, Revision 1 (N RC, 2017 a), and adopted in this document 17 to convert average curvature to average cladding strain, is to assume that the distance from the 18 tensile face of the cladding to the neutral axis is equal to the outside radius, r.

19 Average cladding tensile stress, , should be calculated directly from average cladding strain 20 using the following equation: 21

• E c (Eq. 2-3) 22 Use Equation 2-3 provides a consistent and compatible relationship between stress and strain.

23 2-102.3.3 Calculation of Cladding Strain Using Factored Cladding-Only Properties 1 The following discussion describes a methodology that can be easily implemented to calculate 2 the cladding tensile strain and stress and fuel rod flexural rigidity using only cladding

-only 3 properties. Section 4.2.2 of NUREG/CR

-7198, Revision 1 (N RC, 2017a), presents analyses 4 comparing the measured flexural rigidity from the CIRFT static test results to the calculated 5 flexural rigidity values using the validated cladding-only mechanical property models developed 6 by Pacific Northwest National Laboratory (PNNL) (Geelhood et al., 2008). The purpose of the 7 comparison was to investigate the effect of fuel pellets on the fuel rod's flexural rigidity and 8 cladding strain.

9 Figure 2-7 Comparison of CIRFT Static Bending Results with Calculated PNNL 11 Moment Curvature (Flexural Rigidity) Derived from Claddi ng-Only Stress

-12 Strain Curve (Reproduction of Figure 22 from NUREG/CR-7198, Revision 1 13 (N RC, 2017a)). S1 , S2 , S3, and S4 Represent the Experimental Results for 14 HBR HBU SNF As-Irradiated Specimens, HR2 Represents the Experimental 15 Results for HBR HBU SNF Hydr ide-Reoriented Specimen, and PNNL 16 Represents the Results Calculated Using the Validated Cladding-Only 17 Mechanical Property Models Developed by P NNL (From Geelhood et al., 18 2008) 19 T he CIRFT static test results plotted in Figure 2-7 show the moment

-curvatur e response of the 21 four HBR HBU SNF as-irradiated specimens S1, S2, S3, and S4 and the hydride

-reoriented 22 specimen HR2. The loading portion of the moment

-curvature response begins at 0 N

• m and 23 reaches a maximum at about 80 N

• m, at which point the specimens begin to unload. The 24 moment-curvature responses of the four HBR HBU SNF as

-irradiated specimens during loading 25 were similar up to a moment of 35 N*m. They are characterized by two distinct linear 26 responses, EI1 and EI2, followed by a nonlinear response during the loading and a linear 27 response upon unloading (EI3) (Figure 2-8). 28 2-11Also shown in Figure 2

-7 is the cladding

-only moment

-curvature loading curve constructed 1 using the PNNL cladding-only mechanical property models.

The static test results for both as

-2 irradiated and hydride

-reoriented specimens show much higher bending moment resistance 3 during loading compared to the PNNL cladding-only data. The slopes, EI1 and EI2, of the four 4 HBU fuel rods are greater than the slope of the PNNL data for the cladding-only rod. 5 Figure 2-8 Characteristic Points on Moment-Curvature Curve. A , B , C, a nd D are 7 Points o n the Curve. EI1 i s the Slope of the Loading Curve Between 0 a nd 8 A.EI2 i s the Slope of the Loading Curve Between A a nd B. EI3 i s the Slope9 of the Unloading Curve Between D and 0. The Cladding

-Only Moment

-10 Curvature Loading Curve Constructed Using the PNNL Cladding-Only 11 Mechanical Property Models is not Shown (Reproduction of Figure 21 from 12 NUREG/CR-7 198, Revision 1 (N RC, 2017a))13 Figure 2-7 also shows that at bending moments during loading less than 35 N*m, the flexural 15 rigidities of the four as

-irradiated rods, which have only circumferential hydrides, and HR2, 16 which has both circumferential and radial hydrides, are essentially the same. This result 17 supports the pretest expectation that, because the bending tensile stress in the cladding is 18 parallel to the plane of both the radial and circumferential hydrides, the presence of radial 19 hydrides would not significantly alter the flexural response from the case where only 20 circumferential hydrides are present. The results of tests currently being conducted by the U.S.

21 Department of Energy (DOE) will further confirm this hypothesis as it applies to other cladding 22 types. 23 In the CIRFT static test results for HBR HBU SNF rods shown in Figure 2

-7, no failures 25 occurred. The lower

-bound maximum moment achieved in the tests is approximately 80 N*m. 26 In addition, it is important to point out that a bending moment of 80 N*m is significantly greater 27 than the bending moment an HBR HBU SNF rod will experience during an HAC 9

-m (30-ft) side 28 drop (see Section 2.3.4.1). This means that fuel rod integrity is expected to be maintained 29 during an HAC drop scenarios, and therefore, fuel rod reconfiguration is very unlikely.

30 2-12 For the as

-irradiated HBR HBU SNF rods, Table 2

-1 shows that in the EI1 region of the 1 moment-curvature results, the average flexural rigidity is 2.66 (.e., 71.58 N*m 2/26.93 N*m 2) 2 times greater than the cladding-only case, and in the EI2 region the average flexural rigidity is 3 2.16 (i.e., 58.10 N*m 2/26.93 N*m 2) times greater than the cladding

-only case. For the 4 hydride-reoriented fuel rod, HR2, Table 2

-1 shows that in the EI1 region

, the average flexural 5 rigidity is 2.33 (i.e., 62.77 N*m 2 / 26.93 N*m 2) times greater than the cladding

-only case, and in 6 the EI2 region

, the average flexural rigidity is 1.54 (i.e., 41.52 N*m 2 / 26.93 N*m 2) times greater 7 than the cladding

-only case.

8 Table 2-1 Comparison of Average Flexural Rigidity Results Between CIRFT Static 9 Testing a nd PNNL Cladding-Only Data (From Validated Mechanical 10 Property Models in Geelhoo d e t al., 2008) 11 EI1 (N*m 2) EI2 (N*m 2) EI3 (N*m 2) As-Irradiat ed (S1, S2, S3

, and S4) 71.576 58.099 48.133 Hydride-Reoriented (HR2) 62.769 41.517 43.333 Cladding-Only (validated PNNL models) 26.933 - - Table 2-2 Characteristic Points and Quantities Based on Moment-Curvature Curves 12 (Reproduction , in Part, o f Table 4 from NUREG/CR-7198, Revision 1 13 (NRC , 2017a)) 14 Spec label EI1 2) EI2 2) EI2 2) A (m-1) B (m-1) C (m-1) D (m-1) M A M B M C M D S1 78.655 57.33 51.027 0.202 0.968 2.009 2.166 16.695 60.599 83.595 85.413 S2 73.016 60.848 52.699 0.32 1.009 2.001 2.154 20.18 62.133 85.914 87.294 S3 71.517 59.369 47.101 0.311 0.933 2.149 2.308 22.338 59.288 83.728 85.235 S4 63.117 54.849 41.704 0.503 0.862 2.329 2.507 28.54 48.244 81.656 85.02 As-irradiated Avg. 71.576 58.099 48.133 0.334 0.943 2.122 2.284 21.938 57.566 83.723 85.741 As-irradiated Std. Dev. 6.422 2.603 4.886 0.125 0.062 0.154 0.164 4.977 6.322 1.741 1.048 HR2 62.769 41.517 43.333 0.487 1.007 1.585 2.158 30.301 51.884 66.809 79.606 In developing a simplified methodology using cladding

-only mechanical properties

, the staff 15 considers it conservative to use the flexural rigidity ratio from the EI2 data. More specifically, 16 using the average minus two standard deviations of the EI2 data from Table 2-2 is 52.90 N*m 2 17 (i.e., 58.10 2 - 2 (2.60 2)), which results in an EI2 ratio of an HBU fuel rod to a 18 cladding-only rod of 1.96 (i.e., 52.90 2 / 26.93 2). The average minus two standard 19 deviations has a 98 percent exceedance probability, which means there is a 98 percent chance 20 that the actual value of the EI ratio will be greater than 1.96. To account for the effects of 21 hydride reorientation, this result is reduced by 0.713 (i.e., 1.54/2.16), which is the ratio of the 22 2-13 reoriented hydride results to the as

-irradiated resul ts that were calculated in the previous 1 paragraph. Multiplying 1.96 by 0.713 results in a factor of 1.40. However, recognizing the 2 limited test data available to calculating the 1.40 factor, the factor has been further reduced to 3 1.25 to account for the additional uncertainty associated with using limited data. Thus, for the 4 purpose of calculating lateral displacements in the simplified methodology, the flexural rigidity of 5 the HBU fuel rod is equal to the flexural rigidity of the cladding

-only rod multiplied by the factor 6 1.25: 7 (EI)HBU rod = 1.25 (EI)clad only (Eq. 2-4) 8 The curvature, , of the HBU fuel rod is given by: 9 = M/(EI)HBU rod (Eq. 2-5) 10 or: 11 = M/[1.25 * (EI)clad only] (Eq.2-6) 12 where M is the bending moment in the rod.

13 The tensile strain is given by: 14 =

• ymax (E q. 2-7) 15 where ymax is equal to the outer radius, r, of the rod, and the maximum equivalent tensile stress 16 is given by: 17 =

• E c (Eq. 2-8) 18 The methodology described above for using cladding-only properties to calculate cladding 19 strains while accounting for the increased flexural rigidity imparted by the fuel pellet can also be 20 applied to cladding alloys other than Zircaloy-4. Once CIRFT static bending results for other 21 HBU SNF rods (i.e., ZIRLO

-clad and M5

-clad rods) are obtained under planned DOE-sponsored 22 research (Hanson et al

., 2016), this methodology can be replicated to obtain a numerical factor 23 that allows for crediting the flexural rigidity of the fuel pellet in those fuel types. Until those 24 results are available, t he staff considers the use of claddin g-only mechanical properties to 25 calculate cladding stress and strain to be conservative. The staff expects that CIRFT static 26 bending results for other HBU SNF rods obtained by the DOE-sponsored research will confirm 27 this conclusion.

28 2.3.3.1 Two Alternatives for Calculating Cladding Stress and Strain During Drop 29 Accidents 30 Two alternatives can be used to calculate cladding stress and strain, and cladding flexural 31 rigidity, for the evaluation of drop accident scenarios. The first alternative is to use cladding

-32 only mechanical properties from as

-irradiated cladding (which has only circumferential hydrides

) 33 or from hydride-reoriented cladding (which would account for radial hydrides precipitated after 34 the drying process). As discussed in Section 2.3.3, the staff considers that the orientation of the 35 hydrides is not a critical consideration when evaluating the adequacy of cladding

-only 36 mechanical properties. The properties necessary to implement this alternative are derived from 37 cladding-only uniaxial tensile tests and include modulus of elasticity, yield stress, ultimate 38 2-14 tensile strength and uniform strain, and the strain at failure (i.e., the elongation strain).

1 Additional considerations for acceptable cladding

-only mechanical properties (i.e., alloy type , 2 burnup, an d temperature) may be found in either of the current standard review plans (SRPs) 3 for dry storage of SNF (NUREG-1536, Revision 1, "Standard Review Plan for Spent Fuel Dry 4 Storage Systems at a General License Facility

," issued in July 2010 (NRC, 2010) for the review 5 of applications for Certificates of Compliance under 10 CFR Part 72

and NUREG-1567, 6 "Standard Review Plan for Spent Fuel Storage Facilities," issued in March 2000 (NRC, 2000 a) 7 for the review of applications for specific licenses under 10 CFR Part 72) or transportation 8 (NUREG-1617, "Standard Review Plan for Transportation Packages for Spent Nuclear Fuel,"

9 issued in March 2000 (NRC, 2000b))

- hereafter these documents will be referred to as the 10 current SRPs for dry storage or transportation for SN F.2 11 The second alternative is to use cladding

-only mechanical properties that have been modified 12 by a numerical factor to account for the increased flexural rigidity imparted by the fuel pellet.

13 This numerical factor is obtained from static CIRFT static bending results for fully-fueled rods for 14 the particular HBU SNF cladding type and fuel type, as previously discussed. However, this 15 second alternative would be necessary only if the structural evaluation using cladding

-only 16 mechanical properties is unsatisfactory. The acceptance criteria for cladding performance 17 following dry storage and transport

-related drop accident scenarios can be found in the current 18 SRPs for dry storage and transportation of SNF, respectively.

19 2.3.4 Applicability t o Dry Storage a nd Transportation 20 Argonne National Laboratory defined the radial hydride continuity factor (RHCF) as the ratio of 21 the maximum length of continuous radial

-circumferential hydrides projected in the radial 22 direction to the cladding thickness within a 150

-m arc length (see Section 1.5.4). This metric 23 can be used to quantify the degree of reorientation induced in the hydride-reoriented specimen 24 that was static

-bend tested in the CIRFT instrument (specimen HR2). Figure 2-9 shows a 25 metallographic image of the hydride microstructure of test specimen HR1 (used for CIRFT 26 dynamic testing) after the aggressive hydride reorientation procedure used for HBR HBU SNF 27 rod specimens

.3 The HR2 specimen underwent the same radial hydride treatment (Figure 2-10) 28 as HR1, which is considered to be conservative relative to the conditions expected during drying 29 and short-term loading operations (i.e.

, bounding cladding temperature and hoop stresses, 30 multiple thermal cycling).

4 31 During the radial hydride treatment, each specimen was pressurized to induce a maximum hoop 32 stress of 140 MPa at a target temperature of 400 °C for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />, cooled at 1

°C/min to 170

°C, 33 and then heated at 1

°C/min to the hold temperature of 400

°C. This thermal cycling was 34 repeated for five cycles (a condition that HBU SNF assemblies would not experience in practice, 35 2 The current SRPs for dry storage of SNF are being consolidated into a single document, NUREG

-2215, "Standard Review Plan for Spent Fuel Dry Storage Systems and Facilities," Draft Report for Comment issued November 2017 (NRC, 2017b). Similarly, the current SRP for transportation of SNF is being revised and will be reissued as NUREG

-2216, "Standard Review Plan for Approval of Transportation Packages" (NRC, 2018). Both documents will incorporate current Interim Staff Review Guidance documents. The new SRPs will be issued for public comment and are expected to be finalized prior to final issuance of this report.

3 Section 3.4.1 of NUREG

-7198, Revisi on 1 (NRC, 2017a), presents a more detailed discussion o f the radial hydride treatment used for preparation of the Phase II specimens.

4 The cladding alloy in the HR2 test specimen (Zircaloy

-4) had a hydrogen content considered representative for cold-worked, stress

-relieved alloys (Zircaloy

-4, ZIRLO) and considered bounding for recrystallized

-annealed alloys (Zircaloy

-2, and M5).

2-15 if drying operations are performed according to the guidance in ISG

-11, Revision 3, "Cladding 1 Considerations for the Transportation and Storage of Spent Fuel," issued November 2003 2 (NRC, 20 03 a)-see prior Section 1.2 of this report) to further induce a higher fraction of radial 3 hydrides. The specimen was then furnace cooled from 170

°C to room temperature after the 4 last cycle and the pressure was released

. 5 The conservative conditions chosen for the radial hydride treatment are evidenced by the high 6 radial hydride fraction observed after metallography following testing. As Figure 2

-9 shows, the 7 conservative conditions of the radial hydride treatment induced a near-100-percent RHCF in 8 some sections of the rodlet specimen

. Since the radial hydride treatment produced the highest 9 degree of reorientation that could be anticipated, it is therefore expected to provide the most 10 limiting mechanical properties.

11 Figure 2-9 High Magnification Micrograph Showing Radial Hydrides of a HBR HBU 12 SNF Hydride-Reoriented Specimen Tested Under Phase II (Specimen HR1 13 Results Shown; Hydrogen Content 360-400 Wppm) (Reproduction of 14 Figure 35a in NUREG/CR-7198, Revision 1 (N RC , 2017a)) 15 2-16 Figure 2-10 Representative Conditions Used for Radial Hydride Treatment for 1 Preparation of HBR HBU SNF Hydride

-Reoriented Specimens Tested Under 2 Phase II.

The HBU SNF Specimen Wa s Pressurized to 140 Mpa at 400 C 3 With Five Thermal Cycles (Reproduction Of Figure 14 from NUREG/CR

-4 7198, Revision 1 (N RC, 2017a)) 5 The static test results for the hydride

-reoriented Zircaloy

-4 fuel rod (specimen HR2; Figure 2

-7) 6 show minimal difference in the flexural response compared to the as

-irradiated rods up to the 7 bending moments pertinent to a 9

-m (30-ft) drop accident (i.e., bending moments below 35 N

• m 8 - see Section 2.3.4.2 for pertinent calculation). More importantly, the flexural rigidity of the 9 hydride-reoriented specimen is still markedly higher than the calculated cladding-only response 10 according to validated PNNL mechanical property models. The major difference between the 11 response of the hydride-reoriented HR2 specimen and the as

-irradiated rods is the slightly lower 12 flexural resistance of HR2 at higher loads. The slightly lower flexural resistance at higher loads 13 may be the result of the higher density of hydrides in HR2 or the greater extent to which 14 debonding occurred between the cladding and pellet away from the pellet

-to-pellet crack 15 interface.

However, those loads would not be expected during transportation or dry storage 16 operations.

17 The static test results for the hydride

-reoriented HR2 and the as

-irradiated HBR HBU SNF 18 Zircaloy-4-clad fuel rods support the staff's conclusion that the use of cladding

-only mechanical 19 properties is adequate for the structural evaluation of HAC and NCT drop events. Further, t he 20 HAC drop events required for transportation packages apply inertia loads to the fuel rods that 21 bound the design basis storage drops (e.g., drops during transfer operations and no n 22 mechanistic tip over). Therefore, this conclusion based on the CIRFT static test results of 23 Zircaloy-4 can be applied to both transportation and storag

e. 24 The cladding strains that control the static response of an intact fuel rod are the high tensile 25 strains at the face of the crack at the pellet

-pellet interface.

If a pinhole or hairline crack were to 26 be present at this location

, it could have an effect on the static test results because of the strain 27 concentrations they may create. However, the staff considers the probability that a pinhole or 28 hairline crack is at the pellet

-pellet crack face simulateneously longitudinally and 29 2-17 circumferentially to be low. Therefore, it is reasonable that the CIRFT static test results for 1 intact fuel rods can also be applied to undamaged fuel with pinholes or hairline cracks.

2 Since the hydride

-reoriented Zircaloy

-4 fuel rod had a nearly 100 percent RHCF, the staff 3 considers that the same response should be observed by all modern commercial cladding alloy 4 types that may experience hydride reorientation (i.e., Zircaloy-2, ZIRLO and M5). The staff has 5 also reviewed proprietary and non

-proprietary data on end

-of-life rod internal pressures for fuel 6 rods with boron

-based integral fuel burnable absorbers (see Section 1

.5.3) and considers that 7 these rods are also reasonably bound by the maximum rod internal pressure used in the CIRFT 8 radial hydride treatment (i.e., 140 MPa). The staff's expectation is that future DOE

-sponsored 9 CIRFT static testing conducted on other cladding alloy types will provide confirmation of this 10 conclusion (Hanson et al

., 2016). 11 2.3.4.1 Use o f Static Test Results to Evaluate Safety Margins in an HAC Side Drop 12 Event 13 The CIRFT static test results can be used to determine a lower bound safety margin against fuel 14 rod failure during a n HAC side drop event. The safety margin is calculated by dividing the load 15 (or moment) at rod failure by the maximum applied load (or moment) occurring during the side 16 drop event.

17 Figure 2-7 shows that static testing of the HBR HBU SNF rods did not result in rod failures. The 18 lower bound maximum moment achieved in the tests is approximately 80 Nm. Based on the 19 slope of the curves at 80 Nm, it is reasonable to assume that rod failure probably occurs at a 20 moment at or below 100 Nm. Therefore, using 80 Nm provides a conservative basis for 21 calculating safety margin. To quantify the safety margin it is necessary to know the bending 22 moment in the fuel rod as a function of the g

-load acting on the rod due to a side drop event.

23 Each fuel rod in the fuel assembly is supported by grid spacers at multiple locations along the 24 rod. Therefore, for the purpose of calculating the maximum bending moment, the rod can be 25 idealized as a uniformly loaded continuous beam.

26 Relationship Between Applied G-Load and Bending Moment 27 For the purpose of evaluating a safety margin, two different fuel rods are initially considered.

28 The first is a fuel rod from a PWR 15 x 15 fuel assembly, and the second is a n HBR fuel rod 29 that was tested by ORNL in the CIRFT testing device and reported in NUREG/CR

-7198 , 30 Revision 1 (N RC, 2017 a). 31 The properties of the PWR 15 x 15 rod (Table 2-3) are taken from NUREG

-1864, "A Pilot 32 Probabilistic Risk Assessment of a Dry Cask Storage System at a Nuclear Power Plant,"

33 Appendix C, Table C.1, issued March 2007 (NRC, 2007 a). 34 Table 2-3 Properties of PWR 15 x 15 SNF R od 35 Total fuel r od weight 7.011 lb Fuel length 154 in. Number of grid spacers 8 Rod length between g rid spacers (l) 20.5 in. Uniform applied load (w = 7.011 lb / 154 in

.) 0.0455 lb/in

.

2-18 The maximum moment in a uniformly

-loaded continuous beam can be approximated by the 1 maximum moment in a uniformly loaded three

-span continuous beam as shown in Eqn. 2-9: 2 Mmax = 0.100

• w

• l 2 (Eqn. 2-9) 3 i.e., Mmax = (0.100)(0.0455 lb/in.)(20.5 in.)2 = 1.91 lb*in. = 0.2 16 N*m 4 This is the moment resulting from a 1 g

-loading. The g

-load necessary to produce a moment of 5 1 N*m = 1 g / 0.2 16 N*m = 4.63 g / N*m. 6 For the HBR HBU SNF rod, the weight per unit length is calculated from the weight density of 7 fuel and the weight density of cladding, which can be determined from the information in 8 NUREG-1864 , Table C.1 (NRC, 2007 a) for a BWR 7 x 7 fuel rod.

9 Fuel density = 0.34 lb / in.3 10 (i.e., 9.60 lb / [()(0.25)2(144)] = 0.34) 11 Cladding density = 0.234 lb / in

.3 12 (i.e., 1.98 /

[()(0.535)(0.035)(144)

] = 0.234) 13 The diameter (outer, inner) and thickness of the cladding of a n HBR HBU SNF rod as given in 14 NUREG/CR-7198, Revision 1 (N RC, 2017 a) are: 15 Outer diamete r = 10.743 mm = 0

.423 in. 16 Cladding thickness

= 0.748 mm = 0.0294 in

. 17 Inner diameter

= 0.364 in

. 18 From the HBR HBU SNF rod cross-sectional dimensions and the fuel and cladding densities 19 calculated using the data for the BWR 7 x 7 fuel rods, the fuel and cladding weight per unit 20 length can be calculated as follows:

21 HBR fuel weight = 0.0354 lb / in. 22 HBR cladding weight = 0.0085 lb

/ in. 23 w = 0.0354 + 0.0085 = 0.0439 lb / in. 24 l = distance between HBR SNF assembly grid spacers = 26.2 in

. 25 Mmax = (0.100)(0.0439)(26.

2)2 = 3.01 lb*in = 0.340 N*m 26 This is the moment resulting from a 1 g

-loading. The g

-load necessary to produce a moment of 27 1 N*m = 1 g / 0.340 N*m = 2.94 g / N

• m. 28 This example illustrates the fact that the static transverse g

-load necessary to produce a 29 bending moment of 1 N*m in a fuel rod supported by multiple grid spacers varies from rod to 30 rod. For the two rods in this example, the static transverse g

-load required to produce a 31 2-19 bending moment of 1 N*m varied from 2.9 to 5 g depending on the rod cross sectional 1 dimensions and assembly geometry.

2 2.3.4.2 Dynamic Response of a Fuel Rod 3 During a HAC 9

-m (30-ft) side drop of a transportation package with impact limiters, the cask 4 body will typically experience inertia loads on the order of 50 g. However, the fuel rod is flexible, 5 as are the intervening components that support the rod between the cask body and the rod.

6 Therefore, the rigid body deceleration of the cask body will be amplified during a side drop event 7 by the flexibility of the rod and intervening components, resulting in a g

-load in the rod that is 8 higher than the g

-load acting on the cask body. This increase in g

-load is expressed by a 9 dynamic load factor (DLF), which is the ratio of the deflection due to a dynamically applied load 10 to the deflection that would have resulted from the static application of the load.

The DLF will 11 depend on the rod's natural frequency, the duration of the loading, and the shape of the load 12 time history.

13 Since natural frequency, load duration and load time history shape all depend on the physical 14 characteristics of the fuel assembly, the rod and the cask, including impact limiters, a 15 conservative approach will be used to calculate safety margin by using a maximum DLF of 2.0 16 (Biggs, 1964). 17 Thus, the statically equivalent g

-load the fuel rod is subjected to is 18 (DLF) * (50 g) = 2.0 * (50 g) = 100 g 19 which produces a bending moment in the rod of 20 100 g / (2.94 g/N*m) = 34.0 N*m 21 The safety margin (SM) against fuel rod failure during a side drop event is then 22 SM = (80 N*m)/(34.0 N*m) = 2.35 23 2.3.4.3 Seismic Response o f a Fuel Rod 24 The seismic response of a fuel rod can be determined using a variety of structural models.

25 These range from simple idealized models, for which hand calculation methods could be used, 26 to very detailed finite element models. The seismic loads can be applied to these models using 27 either the response spectrum method or a time history analysis method. However, regardless 28 of whether the fuel rod is in a DSS or transportation package, seismic loads will not dominate 29 fuel rod response, because the g

-loads produced by a seismic event are not large enough. In 30 storage the g

-loads applied to the fuel are dominated by the non

-mechanistic tipover event and 31 in a transportation package the g-loads applied to the fuel rod are dominated by the HAC. Both 32 of these events produce g

-loads on the fuel rod that are approximately an order of magnitude 33 larger that the g

-loads produced by a seismic event. In addition, these two events do not occur 34 coincidently with a seismic event and therefore the seismic event does not add to either of these 35 two events.

36 2-20 2.4 Application of Fatigue Test Results 1 2.4.1 Lower Bound Fatigue S-N Curves 2 Fatigue strength data are commonly presented in the form of an S

-N curve, where S is a 3 strength parameter, such as stress or strain, and N denotes the number of cycles to failure at a 4 specific value of the strength parameter. The objective of this section is to develop a lower 5 bound fatigue S

-N curve, that envelopes the HBR HBU Zircaloy

-4 fuel rod fatigue data and 6 includes both as-irradiated rods and rods with reoriented hydrides.

7 Table 2-4 presents the fatigue test data for the HBR HBU fuel rods. In Figure 2-11 , half of the 8 cladding strain range (which isin Table 2-4 and the maximum strain (//max) are plotte d 9 against the number of cycles required to produce cladding failure at a particular strain 10 amplitude.

The strain range is the average of the strains caused by positive and negative 11 bending moments

, which produce different values of curvature and hence strain. The maximum 12 strain is the maximum of the se two strains.

13 Table 2-4 Summary o f CIRFT Dynamic Test Results for A s-Irradiated and Hydride

-14 Reoriented HBR HBU SNF (Reproduction of Table 6 i n NUREG/CR-7198, 15 Revision 1 (NRC, 2017a)) 16 Spec label Seg. ID Load amp. (N) Moment amp. Number of cycles Failure a (m-1) max m1) a (MPa) a (percent) max (percent) D0 605D1F 250 24.068 2.50E+04 Yes 0.439 0.444 206.109 0.236 0.239 D1 607C4B 150 14.107 1.10E+05 Yes 0.215 0.24 117.26 0.117 0.13 D2 608C4B 50 4.207 6.40E+06 No 0.046 0.067 35.496 0.025 0.036 D3 605C10A 100 9.17 1.00E+06 Yes 0.125 0.171 77.938 0.067 0.092 D4 605D1C 75 6.726 1.10E+07 No 0.089 0.12 57.596 0.048 0.065 D5 605D1B 90 8.201 2.30E+06 Yes 0.114 0.123 69.706 0.061 0.066 D6 609C4 125 11.624 2.50E+05 Yes 0.205 0.218 99.546 0.11 0.117 D7 609C3 200 18.923 6.50E+04 Yes 0.351 0.37 160.835 0.189 0.199 D8 606C3E 87.5 7.743 1.28E+07 No 0.107 0.118 66.309 0.057 0.063 D9 609C7 350 33.667 7.10E+03 Y es 0.576 0.624 288.308 0.31 0.335 D10 606C3A 125 11.552 1.80E+05 Yes 0.174 0.213 98.185 0.094 0.115 D11 607C4A 300 29.021 5.50E+03 Yes 0.469 0.564 241.223 0.254 0.306 D12 608C4A 110 9.986 3.86E+05 Yes 0.144 0.171 83.617 0.078 0.092 D13 606B3E 135 12.55 1 1.29E+05 Yes 0.151 0.199 106.677 0.081 0.107 D14 606B3D 87.5 7.842 2.74E+05 Yes 0.112 0.135 66.652 0.06 0.073 D15 606B3C 75 6.639 2.24E+07 No 0.087 0.125 56.426 0.047 0.067 HR1 607D4C 150 15.152 4.19E+04 Yes 0.424 0.433 128.788 0.228 0.233 HR3 608D4A 100 8.982 2.44E+05 Yes 0.219 0.233 76.342 0.118 0.125 HR4 608D4C 160 14.759 5.47E+04 Yes 0.323 0.344 125.449 0.174 0.185 2-21 Figure 2-11 Plots of Half of the Cladding Strain Range (/2) a nd the Maximum Stra in 1 (//Max) a s a Function of Number of Cycles to Failure.

Markers with Arrows 2 Indicate that the Tests Were Stopped Without Failure. (Reproduction of 3 Figure 31b In NUREG/CR 7198, Revision 1 (NRC, 2017a)) 4 The lower bound enveloping S

-N curve for the HBR HBU SNF rods is composed of three 5 straight line segments when plotted on a linear

-log scale. To account for uncertainty with 6 respect to future test results and the influence of higher test temperatures, the equivalent strain 7 amplitude of all segments has been reduced by a factor of 0.9. The 0.9 is justified to account 8 for uncertainty with respect to future test results particularly at higher temperatures.

Each 9 segment's beginning and end point labels from Table 2-4 coordinates (equivalent strain 10 amplitude percent, number of cycles to failure) are given in Table 2-5 and plotted in Figure 2

-12. 11 Table 2-5 Coordinates for Lower-Bound Enveloping S-N Curve for the HBR HBU SNF 12 Rods (Equivalent Strain Amplitude Percent, Number of Cycles to Failure) 13 Segment Beginning Point End Point 1 (D11 to D13)

(0.275, 5.50E+3)

(0.096, 1.29E+5) 2 (D13 to D14)

(0.096, 1.29E+5)

(0.066, 2.74E+5) 3 (D14 to D15)

(0.066, 2.74E+5)

(0.06, 2.4E+)

2-22 Figure 2-12 CIRFT Dymanic (Fatigue) Test Results for As

-Irradiated and Hydride

-1 Reoriented H.B. Robinson Zircaloy

-4 HBU Fuel Rods. The Calculated 2 Lower-Bound Fatigue Endurance Curve is also Shown 3 Fatigue data for reoriented cladding alloys other than Zircaloy

-4 (i.e., Zircaloy

-2, ZIRLO and M5) 4 are not yet available. However, the staff believes the methodology described above for 5 developing a lower

-bound fatigue curve can be used to construct a lower

-bound fatigue curve 6 for other cladding alloys once the as

-irradiated fatigue data become available. The fatigue data 7 plotted in Figure 2

-11 show that at the same number of cycles all of the Zircaloy

-4 fuel rods with 8 reoriented hydrides failed at nearly the same strains as the as

-irradiated Zircaloy-4 fuel rods. 9 Rod specimen D2, which did not fail, was tested at a very low moment amplitude resulting in a 10 very low maximum strain amplitude. The test was also terminated prematurely at 6.4 x 10 6 11 cycles. Based on the results for the other test specimens that did not fail, it would be expected 12 that specimen D2 would not have failed until 1 x 10 8 cycles or beyond. Therefore, rod specimen 13 D2 is not included in the development of the lower bound curve since it would have 14 inappropriately skewed the results. Therefore, the staff considers that a lower

-bound fatigue 15 curve developed from as

-irradiated data for other cladding alloys is adequate for assessing the 16 fatigue life of alloys with reoriented hydrides. Additional fatigue data for hydride

-reoriented 17 specimens for other cladding alloys to be obtained under DOE

-sponsored research are 18 expected to confirm these expectations.

19 2.4.2 Fatigue Cumulative Damage Model 20 During NCT if a fuel rod were to vibrate at a constant strain amplitude, all that would be 21 necessary to predict the fatigue life of the rod is the S

-N curve. However, fuel rod vibration 22 during NCT is expected to have a series of many cycles encompassing a range of strain 23 amplitudes and with each cycle, damage to the fuel rod cladding is continuously accumulating.

24 A fatigue damage model can be used to express how damage from these cycles accumulates.

25 To date, more than 50 fatigue damage models have been proposed, but unfortunately none of 26 these models enjoys universal acceptance, and the applicability of each model varies from case 27 to case. Unlike the aerospace industry, which has conducted extensive research on the 28 2-23 accumulation of fatigue damage to materials, such as steel, aluminum, and titanium, no 1 research has been conducted on fatigue damage to HBU spent fuel cladding. Nevertheless, for 2 many metals, the simple linear damage rule developed by Miner (Gaylord and Gaylord, 1979) 3 appears to provide a simple and reasonably reliable prediction of fatigue behavior under random 4 loadings, and therefore, will be used to evaluate fatigue damage accumulation in HBU SNF rods 5 during NCT.

6 For failure, the linear damage rule is, the following:

7 i = n i/N i = n 1/N 1 + n 2/N 2 + n 3/N 3 + ... = 1 (Eqn 2-9) 8 where: 9 n i = number of strain cycles at strain level i 10 N i = number of strain cycles to produce failure at i. 11 To apply this simple linear damage rule it is assumed that the NCT loading history can be 12 reduced to a series of different strain levels where the number of cycles associated with each 13 strain level, i, is, n.. To account for uncertainty in using a simple linear damage rule to describe 14 the accumulated fatigue damage in HBU fuel, the right side of the above equation should be set 15 equal to 0.7. This value is considered an approximate lower bound for the uncertainty in Miner's 16 damage model (Hashin, 1979).

17 2.4.3 Applicability t o Storage a nd Transportation 18 The CIRFT fatigue tests were conducted under conditions that produced a uniform bending 19 moment in the fuel rod. Thus, these results apply only to loading conditions that produce 20 longitudinal bending stresses in the cladding of the fuel. Such loading conditions occur when 21 fuel rods vibrate during NCT.

Fluctuating loads can also occur during storage when the 22 cladding experiences thermal cycles because of daily and seasonal fluctuations in ambient 23 temperature. These thermal cycles will induce cyclic stresses on the cladding due to changes in 24 fission and decay gas pressure

, which will result in fluctuation s in cladding hoop stresses. As 25 explained above, however, the fatigue test results apply only to loading conditions that produce 26 longitudinal bending stresses in the cladding of the fuel. The fatigue test results are not 27 applicable to loading conditions that produce fluctuations in hoop stress. Therefore, the fatigue 28 test results cannot be applied to thermal fatigue during storage.

29 In the CIRFT static and fatigue tests the fuel rods were subjected to a constant bending moment 30 which resulted in a longitudinal bending stress in the cladding. However, in an actual spent fuel 31 rod there is internal gas pressure, which creates hoop stress on the order of 100 MPa

- see 32 Section 1.5.3.3. The presence of the hoop stresses creates a non

-proportional biaxial stress 33 state in the cladding. The stress state is non

-proportional because the hoop stress remains 34 constant while the longitudinal bending stress fluctuates. Recent research on the effect of 35 proportional biaxial stress fields on fatigue crack growth show s no significant effect of the biaxial 36 stress field on fatigue crack propagation behavior (Pickard , 2015). It is expected that the same 37 result would also hold for non-proportional biaxial stress fields. Based on these results, the staff 38 considers that the presence of a biaxial stress field in a spent fuel rod does not need to be 39 considered Therefore, only the longitudinal bending stresses in the cladding need to be 40 considered when using the ORNL static and fatigue test data.

41 2-24 During storage or transportation, it is possible that a seismic event could occur. Typically the 1 strong motion duration of a seismic event is approximately 10 seconds. A fuel rod generally 2 responds to seismic input in the 10 to 30 h ertz (Hz) frequency range. This means that the 3 number of fatigue cycles associated with a seismic event would be no more tha n about 300 4 cycles (10 seconds x 30 Hz = 300 cycles). In addition, it is expected that the seismic load 5 applied to the rod would be less than 10

-g. Based on the results summarized at the end of 6 Section 2.3.4.1, a 10

-g load would produce a bending moment in the rod of about 3.5 N*m. 7 From Table 2

-4, a bending moment of 3.5 N*m would result in a maximum cladding strain of 8 about 0.03%. From an event that produced 300 bending cycles at a maximum strain of 9 0.03%, Figures 2

-11 and 2-12 show that virtually no fatigue damage would be expected.

10 Therefore, seismic events during storage or transportation are not expected to compromise 11 the fuel integrity.12 3-1 3 DRY STORAGE OF HIGH BURNUP SPENT NUCLEAR FUEL 1 3.1 Introduction 2 The U.S. Nuclear Regulatory Commission (NRC) staff (the staff) has developed example 3 licensing and certification approaches for dry storage of high burnup (HBU) spent nuclear fuel 4 (SNF). Applicants may use these approaches to provide reasonable assurance of compliance 5 with Title 10 of the Code of Federal Regulations (10 CFR) Part 72, "Licensing Requirements for 6 the Independent Storage of Spent Nuclear Fuel, High

-Level Radioactive Waste, and Reactor 7 Related Greater Than Class C Waste,"

during normal, off

-normal and accident conditions of 8 storage. The staff developed these example approaches according to the conclusions of the 9 engineering assessment in Chapter 2. Figure 3-1 provides a hig h-leve diagram of these 10 approaches, which vary based on (1) the condition of the fuel (undamaged or damaged), and 11 (2) the length of time the fuel has been in dry storage. Section 3.2.2. discusses considerations 12 for additional analyses expected for non

-leaktight dry storage system (DSS) designs. An 13 applicant may consider and demonstrate other approaches that may be acceptable. 14 As required by 10 CFR 72.24(b) and 10 CFR 72.236(a), an application for a specific license for 15 an independent spent fuel storage installation (ISFSI) or an application for a Certificate of 16 Compliance (CoC) for a DSS design, respectively, should identify the allowable SNF contents 17 and condition of the assembly and rods per the design bases

. The allowable cladding condition 18 for the S NF contents is generally defined in the Technical Specifications of the specific license 19 (10 CFR 72.44(c)) or CoC (10 CFR 72.

236 (a)), and the nomenclature may vary between 20 different DSS designs. For example, the terms "intact" and "undamaged" have both be en 21 historically used to describe cladding without any known gross cladding breaches. In 22 accordance with 10 CFR 72.212(a)(1) and 10 CFR 72.212(b)(3), users of DSSs (general 23 licensees) are to comply with the Technical Specifications of the CoC by selecting and loading 24 the appropriate fuel, and are to maintain records that reasonably demonstrate that loaded fuel 25 was adequately selected, in accordance with their approved site procedures and Quality 26 Assurance Program. 27 Interim Staff Guidance (ISG)-1 , Revision 2, "Classifying the Condition of Spent Nuclear Fuel for 28 Interim Storage and Transportation Based on Function," issued in May 2007 (NRC, 2007b

), 29 provides guidance for developing the technical basis supporting the conclusion that the SNF 30 (both rods and assembly) to be loaded in a DSS are intact or undamaged.

1 This would include 31 considering whether the material properties, and possibly the configuration, of the SNF 32 assemblies can be altered during the requested dry storage period. If th e alteration is 33 significant enough to prevent the fuel or assembly from performing its intended functions, then 34 the fuel assembly should be classified as damaged.

35 Damaged SNF is generally defined in terms of the characteristics needed to perform functions 36 to ensure compliance with fuel-specific and DSS-related regulations. A fuel

-specific regulation 37 defines a characteristic or performance requirement of the SNF assembly. Examples of such 38 regulations include 10 CFR 72.122(h)(1) and 10 CFR 72.122(l). A DSS-related regulation 39 defines a performance requirement placed on the fuel so that the DSS can meet its regulatory 40 requirements. Examples of such regulations include 10 CFR 72.122(b) and 10 CFR 72.124(a).

41 1 The current revisions of all ISG documents will be rolled into revised standard review plans (SRPs) for dry storage and transportation, as appropriate, and will then be removed from the public domain. The revised SRPs will be issued for public comment prior to being finalized.

3-2The glossary in this report provides the staff's definitions of intact, undamaged, and damaged 1 fuel. For additional information, refer to the current standard review plans (SRPs) for dry 2 storage of SNF (NUREG

-1536, Revision 1, "Standard Review Plan for Spent Fuel Dry Storage 3 Systems at a General License Facility

," issued in July 2010 (NRC, 2010) for the review of 4 applications for Certificates of Compliance under 10 CFR Part 72, and NUREG

-1567, "Standard 5 Review Plan for Spent Fuel Storage Facilities," issued in March 2000 (NRC, 2000a) for the 6 review of applications for specific licenses under 10 CFR Part 72) - hereafter, these documents 7 will be referred to as the current SRPs for dry storage SNF.

2 8 2 The current SRPs for dry storage of SNF are being consolidated into a single document, NUREG

-2215, "Standard Review Plan for Spent Fuel Dry Storage Systems and Facilities," Draft Report for Comment issued November 2017 (NRC, 2017b), which will incorporate current Interim Staff Review Guidance documents. NUREG-2215 has been issued for public comment and is expected to be finalized prior to final issuance of this report.

3-3 Figure 3-1 Example Licensing and Certification Approaches for Dry Storage of High Burnup Spent Nuclear Fuel

3-4 Consistent with the guidance in (ISG)-1 , Revision 2 (NRC, 2007b

), HBU SNF assemblies with 1 any of the following characteristics, as identified during the fuel selection process, are generally 2 classified as damaged unless an adequate justification is provided for not doing so

3 There is visible deformation of the rods in the HBU SNF assembly. This does not refer 4 to the uniform bowing that occurs in the reactor; instead, this refers to bowing that 5 significantly opens up the lattice spacing.

6 Individual fuel rods are missing from the assembly. The assembly may be classified as 7 intact or undamaged if the missing rod(s) do not adversely affect the structural 8 performance of the assembly, or radiological and criticality safety (e.g., there are no 9 significant changes to rod pitch). Alternatively, the assembly may be classified as intact 10 or undamaged if a dummy rod that displaces a volume equal to, or greater than, the 11 original fuel rod is placed in the empty rod location.

12 The HBU SNF assembly has missing, displaced, or damaged structural components 13 such that either:

14 - Radiological and/or criticality safety is adversely affected (e.g., significant change 15 in rod pitch), 16 - The structural performance of the assembly may be compromised during normal, 17 off-normal , and accident conditions of storage, or 18 - The assembly cannot be handled by normal means (i.e., crane and grapple), if 19 the design bas es relies on ready retrieval of individual fuel assemblies.

20 Reactor operating records or fuel classification records indicate that the HBU SNF 21 assembly contains fuel rods with gross rupture

. 22 The HBU SNF assembly is no longer in the form of an intact fuel bundle (e.g., consists 23 of, or contains, debris such as loose fuel pellets or rod segments).

24 Defects such as dents in rods, bent or missing structural members, small cracks in structural 25 members, and missing rods do not necessarily render an assembly as damaged, if the intended 26 functions of the assembly are maintained

i.e., the performance of the assembly does not 27 compromise the ability to meet fuel

-specific and DSS-related regulations.

28 3.2 Uncanned Fuel (Intact and Undamaged Fuel) 29 Undamaged HBU SNF can be stored in the D SS without the need for a separate fuel can (i.e.

, a 30 separate metal enclosure sized to confine damaged fuel particulates) to maintain a known 31 configuration inside the DSS confinement cavity. This fuel includes rods that are either intact 32 (i.e., no breaches of any kind) or that contain small cladding defects (i.e.

, pinholes or hairline 33 cracks) that may permit the release of gas from the interior of the fuel rod. Cladding with gross 34 ruptures that may permit the release of fuel particulate s may not be considered undamaged. The 35 configuration of undamaged HBU SNF may be demonstrated to be maintained if loading and 36 transport operations are designed to prevent and/or mitigate degradation of the cladding and 37 other assembly components, as discussed in ISG-22 , "Potential Rod Splitting Due to Exposure to 38 an Oxidizing Atmosphere during Short

-Term Cask Loading Operations in LWR or Other Uranium 39 Oxide Based Fuel," issued May 2006 (NRC, 2006).

40 3-5 Following the approaches delineated in Figure 3-1, an application for dry storage of undamaged 1 HBU SNF would include a structural evaluation of the fuel rods under design

-bases drop 2 accident scenarios. The evaluation serves to demonstrate that the uncanned fuel remains in a 3 known configuration after a drop accident scenario.

4 Two alternatives may be used to calculate cladding stress and strain, and cladding flexural 5 rigidity, for the aforementioned evaluation of drop accident scenarios. The first alternative, 6 shown in Figure 3-2, is to use cladding

-only mechanical properties from as

-irradiated cladding 7 (i.e., cladding with circumferential hydrides, primarily), or hydride

-reoriented cladding (i.e., 8 cladding that account s for radial hydrides precipitated after the drying process).

9 Figure 3-2 First Approach for Evaluating Design

-Bases Drop Accidents During Dry 10 Storage 11 As discussed in Section 2.3.3, the staff considers the orientation of the hydrides not to becritical 12 when evaluating the adequacy of cladding

-only mechanical properties.

Therefore, the properties 13 necessary to implement this first alternative may be derived from cladding

-only uniaxial tensile 14 tests and include modulus of elasticity, yield stress, ultimate tensile strength and uniform strain, 15 and the strain at failure (i.e., the elongation strain). Refer to the current SRPs for dry storage of 16 SNF for additional considerations for acceptable cladding

-only mechanical properties (i.e., alloy 17 3-6 type, burnup, and temperature) and the acceptance criteria for cladding performance during dry 1 storage operations.

2 A second alternative , shown in Figure 3.3, is to use cladding

-only mechanical properties that 3 have been modified by a numerical factor to account for the increased flexural rigidity imparted 4 by the fuel pellet. This numerical factor is obtained from static test data from the cyclic 5 integrated reversible

-bending fatigue tester (CIRFT) for fully-fueled rods for the particular 6 cladding type and fuel type (see Section 2.3.3

). However, this second alternative would be 7 necessary only if the structural evaluation using cladding

-only mechanical properties is 8 unsatisfactory. Refer to the current SRP for dry storage of SNF for acceptance criteria on 9 cladding performance during dry storage operations

. 10 Figure 3-3 Second Approach for Evaluation of Design-Bases Drop Accidents During 11 Dry Storage 12 3.2.1 Leaktight Confinement 13 Consistent with the guidance in the current SRPs for dry storage of SNF, an application for a 14 DSS for HBU SNF is expected to define the maximum allowable leakage rate for the entire 15 confinement boundary. The maximum allowable leak age rate is based on the quantity of 16 radionuclides available for release and is evaluated to meet the confinement requirements for 17 3-7 maintaining an inert atmosphere within the DSS confinement cavity and compliance with the 1 regulatory limits of 10 CFR 72.104, "Criteria for Radioactive Materials in Effluents and Direct 2 Radiation from an ISFSI or MRS,"

and 10 CFR 72.106, "Controlled Area of an ISFSI or MRS."

3 L eak age rate testing is performed on the entire confinement boundary (over the course of 4 fabrication and loading) and ensures that the package can maintain a leak rate below the 5 maximum allowable leakage rate per ANSI N14.5 (2014). 6 If the entire DSS confinement boundary, including its closure lid, is designed and tested to be 7 "leaktight" as defined in American National Standards Institute (ANSI) N14.5

- 2014, "American 8 National Standard for Radioactive Materials

-Leakage Tests on Packages for Shipment" and the 9 current SRPs for dry storage of SNF, then the application is not expected to include additional 10 dose calculations based on the allowable leakage rate that demonstrate compliance with the 11 regulatory limits of 10 CFR 72.104(a) and 10 CFR 72.106(b). In addition , the structural analysis 12 of the package is to demonstrate that the confinement boundary will not fail under the postulated 13 drop scenarios and that the confinement boundary will remain leaktight under all conditions of 14 storage. Refer to the current SRPs for dry storage of SNF for additional guidance on 15 demonstrating compliance with the leaktight criterion.

16 3.2.2 Non-Leaktight Confinement 17 For those DSS designs not tested to a "leaktight" confinement criterion, the application is 18 expected to include dose calculations based on the allowable leakage rate to demonstrate 19 compliance with the regulatory limits of 10 CFR 72.104(a) and 10 CFR 72.106(b). L eak age rate 20 testing is performed on the entire confinement boundary (over the course of fabrication and 21 loading) and ensures that the package can maintain a leak rate below the maximum allowable 22 leakage rate per ANSI N14.5 (2014). 23 To determine the dose rate for the confinement boundary, an application for a non

-leaktight 24 DSS is expected to provide a technical basis for the assumed bounding HBU fuel failure rates 25 for normal, off

-normal, and accident conditions of storage. If an application is not able to 26 provide and justify its bounding fuel failure rates, then the fuel failure rates below can be 27 assumed as bounding values for normal, off

-normal, and accident conditions of storage

28 Normal conditions of storage: 1 percent 29 Off-normal conditions of storage: 10 percent 30 Accident conditions of storage: 100 percent 31 Bounding Release Fractions for High Burnup Spent Nuclear F uel 32 HBU SNF fuel has different characteristics than low burnup (LBU) SNF with respect to cladding 33 oxide thickness, hydride content, radionuclide inventory and distribution, heat load, fuel pellet 34 grain size, fuel pellet fragmentation, fuel pellet expansion and fission gas release to the rod 35 plenum [See Appendix C.5 to NUREG/CR-7203, "A Quantitative Impact Assessment of 36 Hypothetical Spent Fuel Reconfiguration in Spent Fuel Storage Casks and Transportation 37 Packages," issued September 2015 (NRC, 2015) for additional details on HBU SNF]. These 38 characteristics may affect the mechanisms by which the fuel can breach and the amount of fuel 39 that can be released from failed fuel rods.

Hence, the staff evaluated open literature on HBU 40 fuel rod failure rates and release fractions of Chalk River unknown deposits (CRUD), fission 41 gases, volatiles, and fuel fines to assist in the review of applications for non

-leaktight 42 3-8 confinement boundaries.

Table 3-1 provides release fractions that may be considered 1 reasonably bounding for HBU SNF.

If the release fractions are not used, justification of the 2 proposed release fractions of the source terms is expected to include an adequate description 3 of burnup for the test specimen, number of tests, collection method for quantification of 4 respirable release fractions, test specimen pressure at the time of fracture, and source 5 collection system (sophisticated enough to gather the bounding respirable release fractions

). 6 Table 3-1 Fractions of Radioactive Materials Available for Release from HBU SNF 7 Under Conditions of Dry Storage (for both Pressurized Water Reactor a nd 8 Boiling Water Reactor Fuels) 9 Variable Normal Conditions Off-Normal Conditions Accident-Fire Conditions Accident-Impact Condit ions Fraction of Fuel Rods Assumed to Fail 0.01 0.1 1.0 1.0 Fraction of Fission Gases Released Due to a Cladding Breach 0.15 0.15 0.15 0.35 Fraction of Volatiles Released Due to a Cladding Breach 3 x 10-5 3 x 10-5 3 x 10-5 3 x 10-5 Mass Fraction of Fue l Released as Fines Due to a Cladding Breach 3 x 10-5 3 x 10-5 3 x 10-3 3 x 10-5 Fraction of CRUD Spalling Off Cladding 0.15 0.15 1.0 1.0 CRUD 10 The average CRUD thickness in HBU SNF cladding has been estimated to be similar to that 11 observed on LBU SNF cladding. A review of data in the literature (NRC, 2000 c; Einziger and 12 Beyer, 2007) indicates that a release (spalling off) of 15 percent of cladding CRUD may be 13 assumed as reasonably bounding to both normal and off

-normal conditions of storage, and a 14 release of 100 percent of the cladding CRUD is conservatively bounding to both postulated fire 15 and impact accidents during storage (NRC, 2014). 16 Fission Gases 17 The NRC's FRAPCON steady

-state fuel performance code has been previously used to assess 18 release fraction s of fission gases during transportation (NRC, 2011). The seven most common 19 fuel designs were evaluated using FRAPCON's modified Forsberg

-Massih model (8 x8, 9x9, 20 and 10x10 fuel for boiling water reactors (BWR s) and 14x14, 15x15, 16x16, and 17 x17 for 21 pressurized-water reactors (PWR s). For each fuel design, a number of different power histories 22 aimed at capturing possible realistic reactor irradiations were modeled. The fission gas content 23 within the free volume of the rods was evaluated for a total of 243 different cases (39 for each of 24 the BWR fuel designs

37 for 14x14 and 16x16 PWR fuel designs, and 26 for 15 x15 and 17x17 25 PWR fuel designs

). A review of the results indicates that a release of 15 percent of fission 26 3-9 gases may be assumed as reasonably bounding to normal conditions of transport scenarios for 1 rod average burnup s up to 62.5 GWd/MTU. The same release fraction may be reasonably 2 assumed for both normal and off

-normal conditions of storage.

3 During a fire accident scenario in storage, the fuel is not expected to reach temperatures high 4 enough that fission gases can diffuse out of the pellet matrix or grain boundaries to the rod 5 plenum. The thermal rupture tests showed that release occurred at higher temperatures than 6 those experience d during a transportation fire accident (NRC, 2000 c). The same behavior is 7 expected during a postulated fire accident condition of storage. Therefore, the same release 8 fraction of 15 percent of fission gases during normal/off

-normal conditions of storage may be 9 assumed to be reasonably bounding to the fire scenario under accident conditions of storage

. 10 In the case of postulated impact accident (drop) scenarios (e.g., during transfer or retrieval 11 operations), the pellet may be conservatively assumed to crumble. In this scenario, fission 12 gases retained within the pellet grain boundaries may be released in addition to those already 13 released from the fuel rod free volume (i.e., from the fuel-cladding gap and plenum). The 14 FRAPFGR model in FRAPCON may be used to predict the location of the fission gases within 15 the fuel pellet (NRC, 2011). The model has been validated with experimental data obtained 16 using an electron probe micro analyzer. The FRAPFGR model was used to calculate the 17 maximum fraction of the pellet-retained fission gases that may be released during a drop 18 impact, which was determined to be 20 percent. Therefore, assuming all fission gases within 19 the pellet gra in boundaries are released, a 35 percent (15 percent + 20 percent) maximum 20 release fraction may be assumed to be reasonably bounding to a postulated accident fire 21 scenario during storage. This value accounts for the 15 percent maximum fission gases 22 released from the fuel rod free volume (as calculated with the modified Forsberg

- Massih model) 23 and the 20 percent maximum fission gases released from the fuel pellet grain boundaries (as 24 calculated with the FRAPFGR model). These release fraction estimates are consistent with 25 previous NRC estimates (NRC, 200 0 c; NRC , 2007; Einziger and Beyer, 2007).

26 Volatiles 27 Mo st of the volatile release fractions originate from cesium

-based compounds in the form of 28 oxides or chlorides (NRC, 2000 c; NRC, 2014). These volatiles exhibit a different release 29 behavior in comparison to fission gases. Volatiles tend to migrate and aggregate at the rim on 30 the outer surface of the fuel pellet during reactor irradiation, which is characteristic of burnups 31 near or exceeding 60 GWd/MTU. The pellet rim is characterized by a fine crystalline grain 32 structure (0.1--0.3 µm or submicron in characteristic size) (Spino et al

., 2003; Einziger and 33 Beyer, 2007), a high porosity that may exceed 25 percent, and a high concentration of actinides 34 relative to the inner pellet matrix

. 35 Sandia National Laboratories assessed the maximum release fraction of volatiles (cesium and 36 other ruthenium-based compounds

) under drop and fire accident scenarios of transportation , 37 and determined it to be 0.003 percent (3x10-5) (NRC, 2000 c). This assessment included 38 modeling and analyses using various data from the literatur e. The volatile release fraction 39 during a fire accident scenario was determined to be lower than the release fraction during a 40 drop accident scenario (NRC, 2014; NRC, 2000 c). Therefore, a volatile release fraction of 41 0.003 percent (3 x 10-5) may be assumed to be reasonably bounding to normal, off

-normal, and 42 accident conditions of storage. This release fraction estimate is also consistent with an 43 independent estimate by Einziger and Beyer (2007).

44 45 3-10Fuel Fines 1 Release fractions from SNF fines during storage and transportation have been previously 2 documented (NRC, 2000 c; NRC, 2007; Benke et al., 2012; NRC, 2014). HBU SNF has a 3 different pellet microstructure than LBU SNF, which is characterized by an inner matrix and an 4 outer pellet rim layer. The thickness of the outer pellet rim layer increases with higher fuel 5 burnup. Therefore, differences in microstructure between the inner pellet matrix and the outer 6 pellet rim should be considered when evaluating release fractions of fuel fines from HBU SNF.

7 Although there is no reported literature on HBU SNF rim fracture as a function of impact energy , 8 other data can be used to indirectly assess the contribution of the rim layer to the release 9 fractions of fuel fines. Spino et al (1996) estimated the fracture toughness of the rim layer from 10 micro-indentation tests. Compared to the inner SNF matrix, the rim layer showed an increase of 11 fracture toughness. The increase of fracture toughness implies a decrease of release fraction. 12 Hirose et al (2015) also discussed results of axial dynamic impact tests simulating accident 13 conditions during transport, which are expected to be bounding to postulated drop scenarios 14 during dry storage. The dispersed particles from pellet breakage following impact were 15 collected and correlated to impact energy. The staff has compared the measured release 16 fraction of fuel fines from Hirose et al (2015) with previous NRC estimates of release fraction 17 versus impact energy for SNF and other brittle materials (depleted UO 2, glass and Synroc) (see 18 Figure 3 of NUREG 1864, "A Pilot Probabilistic Risk Assessment of a Dry Cask Storage System 19 at a Nuclear Power Plant" (NRC 2007)). Based on these analyses, the staff concludes that 20 there is no indication that pellet rim layer contributes to increased release fractions. 21 Since the outer HBU fuel pellet rim does not appear to contribute to additional release fractions, 22 previous NRC estimates for release fractions of fuel fines may continue to be used (NRC, 23 2000c; NRC, 2007; Benke, et al., 2012; Ahn et al., 2012; NRC, 2014). Per the range of 24 estimates in the literature, a release fraction for fuel fines of 0.003 percent (3x10-5) may be 25 assumed to be reasonably bounding to normal, off

-normal, and accident (drop impact) 26 conditions of storage. During a fire accident scenario, fuel oxidation is conservatively assumed 27 to increase the release fraction of fuel fines by a factor of 100 (NRC, 2000 c; Ahn et al 2012). 28 Therefore, a 0.3 percent (3x10-3) release fraction of fuel fines may be assumed as reasonably 29 bounding to fire accident conditions of storage.

30 The staff recognizes that various international cooperative research programs are currently 31 investigating release fractions from HBU SNF. Once those data are available to the public, the 32 staff will review and determine whether the conservative estimates in the above discussion 33 should be revis it ed. 34 3.2.3 Dry Storage U p To 20 Years 35 Section 1.2 discussed t he staff's review guidance for the licensing and certification of dry 36 storage of HBU SNF for a period of up to 20 year

s. The technical basis referenced in that 37 guidance supports the staff's conclusion that creep is not expected to result in gross rupture if 38 cladding temperatures are maintained below 400 °C (752 °F). 39 Chapter 2 also provi ded an asses m e of th e effects of hydri de reorientati on per static and 40 fatigue bending test results on HBU SNF specimens. Those test results provide a technical 41 basis for the staff's conclusion that the use of cladding mechanical properties (with either as- 42 irradiated or hydride- reoriented microstructure) is adequate for the structural evaluation of HBU 43 SNF when evaluating postulated drops during dry storage (e.g., drops during transfer 44 3-11 operations, non

-mechanistic DSS cask tipover). Refer to the current SRPs for dry storage of 1 SNF for staff review guidance on additional considerations for acceptable cladding

-only 2 mechanical properties (i.e., alloy type, burnup, temperature), on acceptable references for 3 cladding mechanical properties and on acceptance criteria for the structural evaluation of the 4 HBU fuel assembly for the drop accident scenarios. As indicated in Figure 3-1, supplemental 5 safety analyses are not expected for HBU SNF in dry storage for periods not exceeding 20 6 years. 7 3.2.4 Dry Storage Beyond 20 Years 8 As indicated in Figure 3-1, to address age

-related uncertainties related to the extended dry 9 storage of HBU SNF (i.e., dry storage beyond 20 years), the application is expected to be 10 supplemented with either results from a surrogate demonstration program or supplemental 11 safety analyses assuming justified hypothetical fuel reconfiguration scenarios. The results from 12 a surrogate demonstration program are meant to provide field

-obtained confirmation that the 13 fuel has remained in the analyzed configuration after 20 years of dry storage. If confirmation is 14 not provided, the safety analyses for the DSS should be supplemented to assume reconfigured 15 fuel. Consistent with the requirements in 10 CFR Part 72, the supplemental information may be 16 provided in either the initial license or CoC application (per 10 CFR 72.4 0(a) and 17 10 CFR 72.238, "Issuance of an NRC Certificate of Compliance"

) or in a renewal application 18 (10 CF R 72.42(a) and 10 CFR 72.240(a))

. 19 The NRC has approved the licensing and certification of HBU SNF for an initial 20

-year-term p er 20 the technical basis in the staff's review guidance, as discussed in Section 1.2. However, t h e 21 staff has recognized that the technical basis is based on short

-term accelerated creep testing 22 (i.e., laboratory scale testing up to a few months), which results in increased uncertainties when 23 extrapolated to long periods of dry storage

- see Appendix D to NUREG-1927, Revision 1 24 (NRC, 2016 b). Although the staff has confidence based on this short

-term testing that creep

-25 related degradation of the HBU fuel will not adversely affect its analyzed configuration for 26 storage periods beyond 20 years, there is no operational field-obtained data to confirm this 27 expectation, as was done in the prior demonstration on LBU fuel described in NUREG/CR

-6745, 28 "Dry Cask Storage Characterization Project

-Phase 1; CASTOR V/21 Cask Opening and 29 Examination," issued September 2001 (NRC, 2001),; and NUREG/CR 6831, "Examination of 30 Spent PWR Fuel Rods after 15 Years in Dry Storage," issued September 2003 (NRC, 2003b).

31 In addition, the staff also acknowledges that while the CIRFT results obtained to

-date (as 32 discussed in Chapter 2) provide an adequate technical basis for assessing the separate effects 33 of hydride reorientation, the results do not account for potential synergistic effects of various 34 physical and chemical phenomena occurring during extended dry storage (e.g., cladding creep, 35 hydride reorientation, irradiation hardening, oxidation, hydriding caused by residual water 36 hydrolysis, etc.

- see NUREG-2214, "Managing Aging Processes in Storage (MAPS) Report,"

37 issued October 2017 (NRC, 2017c) for discussions on these phenomena

). Therefore, the staff 38 considers it prudent to gather and review evidence that HBU fuel in dry storage beyond 20 39 years has maintain ed its analyzed configuration be gathered and reviewed.

40 3.2.4.1 Supplemental Results from Confirmatory Demonstration 41 A demonstration program, like that conducted for LBU SNF (NRC, 2003; NRC , 2001; NRC, 42 2003 b), may be used to confirm the results from separate

-effects testing, which has provided 43 the technical bases for dry storage of HBU SNF beyond 20 years

. 44 3-12 3.2.4.1.1 Initial Licensing or Certification 1 Consistent with 10 CFR 72.42(a) and 10 CFR 72.238, an applicant may request approval for dry 2 storage of HBU SNF for periods up to 40 years. These applications are not required to provide 3 aging management programs (AMPs), as these programs are expected only in renewal 4 applications. Instead, for initial license s and CoC approvals for dry storage beyond 20 years (up 5 to 40 years), the application may describe the activities to obtain and evaluate confirmatory data 6 from a demonstration program under the aegis of a maintenance plan. The maintenance plan 7 would be implemented after the initial 20 years of dry storage. Applicants may refer to 8 Appendices B and D to NUREG-1927, Revision 1 (NRC, 2016 b) when developing the 9 description of activities to assess data from the confirmatory demonstration

. 10 3.2.4.1.2 Renewal Applications 11 Consistent with 10 CFR 72.42(a) and 10 CFR 72.240(a), a renewal application for a specific 12 license or CoC, may describe the activities to obtain and evaluate confirmatory data to be 13 performed under the aegis of an AMP.

Applicants may refer to Appendices B and D to NUREG-14 1927, Revision 1 (NRC, 2016 b) when developing the description of activities to assess data 15 from the confirmatory demonstration

. 16 3.2.4.2 Supplemental Safety Analyses 17 As an alternative approach to a confirmatory demonstration for HBU SNF, an application may 18 supplement the design bases with safety analyses that demonstrate the DSS can still meet th e 19 pertinent regulatory requirements by assuming hypothetical reconfiguration of the HBU fuel 20 contents into justified geometric form

s. This alternative approach would demonstrate that the 21 design-bases fuel, even if reconfigured

, can still meet the 10 CFR Part 72 requirements for 22 thermal, confinement, criticality safety and shielding during normal, off

-normal, and accident 23 conditions.

For renewal applications, a separate license amendment or CoC amendment may 24 be required if the changes in the supplemental safety analyses do not meet the acceptance 25 criteria in 10 CFR 72.48, "Changes, Tests, and Experiments."

. 26 In NUREG/CR-7203 (NRC, 2015), ORNL Oak Ridge National Laboratory (ORNL) evaluated the 27 impact of a wide range of postulated fuel reconfiguration scenarios under non

-mechanistic 28 causes of fuel assembly geometry change with respect to criticality, shielding (dose rates), 29 containment, and thermal. The study considered three fuel reconfiguration categories , which 30 were characterized by either category 1, cladding failure

category 2, rod/assembly deformation 31 without cladding failure

or category 3 changes to assembly axial alignment without cladding 32 failure. Within configurations in both Categor y 1 and Category 2, the study identified various 33 scenarios

34 Category 1: cladding failure 35 - Scenario 1(a): breached rods 36 - Scenario 1(b): damaged rods 37 38 39 3-13 Category 2:

rod/assembly deformation without cladding failure 1 - Scenario 2(a): configurations associated with side drop 2 - Scenario 2(b): configurations associated with end drop 3 Category 3: changes to assembly axial alignment without cladding failure 4 The analyses in NUREG/CR

-7203 (NRC, 2015) considered representative SNF transportation 5 packages, and a range of fuel initial enrichments, discharge burnup values, and decay times.

6 Two package designs were analyzed: a general burnup credit (GBC)

-32 package containing 32 7 PWR fuel assemblies and a GBC

-68 package containing 68 BWR fuel assemblies. Although 8 NUREG/CR-7203 did not evaluate reconfiguration in DSSs, the scenarios and analytical 9 methods may also be applicable to those designs, as the loads experienced during transport 10 conditions (normal, hypothetical accident) are expected to bound those experienced during 11 storage (normal, off

-normal and accident). The results in NUREG/CR

-7203 should not be 12 assumed to be generically applicable as fuel reconfiguration may have different consequences 13 for a DSS design other than the generic models evaluated in the study. However, the following 14 sections discuss considerations in developing supplemental safety analyses for other DSS 15 designs according to the reconfiguration scenarios considered in NUREG/CR

-7203. 16 3.2.4.2.1 Materials and Structural 17 An application relying on supplemental safety analyses based on hypothetical reconfiguration of 18 the HBU SNF contents should provide a structural evaluation for the package and its fuel 19 contents using any of the approaches discussed in Section 3.2. The staff will review the 20 structural evaluation and the assumed material mechanical properties, including any changes 21 due to higher temperatures resulting from fuel reconfiguration, in a manner consistent with the 22 guidance in the current SRP for dry storage of SNF

. 23 3.2.4.2.2 Confinement 24 An application relying on supplemental safety analyses based on hypothetical reconfiguration of 25 the HBU SNF is expected to demonstrate that the DSS design meets the regulatory 26 requirements for confinement if data from a surrogate demonstration program, used for 27 confirmatory demonstration following the guidance in NUREG

-1927, Revision 1 (NRC, 2016 b), 28 is not available before the renewal of the license for previously dry

-stored fuel for periods longer 29 than 20 years

. 30 However, if the thermal, structural, and material analyses , together with aging management 31 activities for the DSS subcomponents supporting confinement (i.e., confinement boundary) 3 , are 32 used to provide assurance that the allowable leak rate is maintained even after hypothetical 33 reconfiguration of the fuel under normal, off-normal and accident

-level conditions, supplemental 34 safety analysis for the confinement performance of the DSS design are not expected. Thermal 35 analyses demonstrate that all DSS subcomponents supporting confinement (i.e., confinement 36 3 Aging management activities may be conducted under either the aegis of an NRC

-approved AMP (for renewal applications) or a maintenance plan (for initial license or CoC applications requesting approval for periods exceeding 20 years).

3-14 boundary) will be able to withstand their maximum operating temperatures and pressures under 1 normal, off

-normal and accident-level conditions

. 2 3.2.4.2.3 Thermal 3 Fuel reconfiguration can affect the efficiency of heat removal from the fuel because of changes 4 in (1) thermo-physical properties of the canister gas space stemming from release of fuel rod 5 inert gas and fission product gases, (2) heat source location within the canister, and (3) changes 6 in flow area (convection), conduction lengths (conduction) and radiation view factors (thermal 7 radiation). As part of a defense-in-depth approach for addressing a ge-related uncertainties for 8 uncanned and undamaged HBU fuel in dry storage beyond 20 years, the thermal analyses 9 would be expected to analyze scenarios for normal, off

-normal , and accident conditions of 10 storage by assuming the fuel may become substantially altered. NUREG/CR-7203 (NRC, 2015) 11 describes the impact on the DSS canister pressure and the fuel cladding and DSS component 12 temperatures for various scenarios of fuel geometry changes. These are examined below. In 13 general, the results in NUREG/CR

-720 3 should not be considered generically applicable. The 14 thermal analyses of the application are expected to consider scenarios discussed in 15 NUREG/CR-7203 to determine consistency in the analytical methods, scenario phenomena, 16 and results

. The thermal analyses are expected to assess the impact of the fuel reconfiguration 17 on the fuel cladding and DSS component temperatures and the canister pressure for the 18 particular DSS design. 19 For Scenario 1(a) in Category 1 (see Section 3.2.4.2)

, the fuel rods are assumed to breach in 20 such a manner that the cladding remains in its nominal geometry (no fuel reconfiguration), but 21 depending on the canister orientation (horizontal or vertical), the release of fuel rod fill gas and 22 fission product gases can cause a change to component peak temperatures. For Scenario 1(b) 23 in Category 1 , for configurations where a n assembly (or assemblies) is represented as a debris 24 pile(s) inside its basket cell, fuel reconfiguration has a larger impact on the component 25 temperatures for the vertical orientation than for the horizontal orientation, but the packing 26 fraction of the debris bed has minor impact on the component temperatures.

For both 27 Scenarios 1(a) and 1(b), release of the fuel rod gaseous contents increases the number of 28 moles of gas and therefore increases the canister pressure. The canister pressure is expected 29 to increase with the increased fuel rod release fractions. 30 For Scenarios 2(a) and 2(b), the fuel rods are assumed to remain intact without gaseous 31 leakage into the canister space. T he changes of the fuel assembly lattice (contraction in 32 Scenario 2(a) and expansion in Scenario 2(b)) could cause either an increase or decrease in 33 the component temperatures of the storage system depending on the initial assembly geometry 34 and whether the storage system relies on convection for heat transfer. In general, scenarios 35 Scenario 2(a) and Scenario 2(b) have minor impact on the fuel cladding and DSS component 36 temperatures and canister pressure. For Category 3 , the fuel rods are assumed to remain intact 37 without gaseous leakage into the canister space, but the axial shifting of the assembly changes 38 the heat source location within the canister. Changes in assembly axial alignment within the 39 basket cells are expected to have minor impact on the component temperatures and the 40 canister pressure.

41 Normal, Off-Normal, and Accident Conditions of Storage 42 Based on the thermal phenomena described above and NUREG/CR

-7203 (NRC, 2015), an 43 approach acceptable to staff would evaluate the impact of Scenarios 1(a) and 1(b) on the 44 canister pressure and the fuel cladding and package component temperatures assuming 45 3-15 rupture of 1 percent, 10 percent and 100 percent of the fuel rods for normal, off

-normal, and 1 accident conditions, respectively.

2 Although Scenarios 2(a) and 2(b) in Category 2 and Category 3 are not expected to have a 3 significant impact on DSS thermal performance under normal, off

-normal and accident 4 conditions, because the fuel rods in Scenarios 2(a), 2(b) and 3 are assumed to remain intact 5 without gaseous leakage into the canister space

, the applicant may need to provide a thermal 6 evaluation due to specifics of the DSS design

. 7 3.2.4.2.4 Criticality 8 An application may demonstrate that a DSS meets the regulatory requirements for criticality 9 safety for the period beyond 20 years by assuming hypothetical reconfiguration of the HBU 10 SNF into a bounding geometric form. This approach is one way to ensure compliance with 11 10 CFR 72.124, "Criteria for Nuclear Criticality Safety,"

or 10 CFR 72.236(c) during normal, off-12 normal, and accident conditions

, if the structural evaluation does not adequately define the 13 mechanical properties of the cladding

. 14 As mentioned previously, ORNL examined hypothetical fuel reconfiguration for various 15 scenarios and the impacts on the criticality safety of a DSS and documented the results in 16 NUREG/CR-7203. This study, considers burnup up to 70 GWd/MTU for criticality evaluations.

17 NUREG/CR-7203 provides some insight into the reactivity trends for various reconfiguration 18 scenarios; howeve r the results in NUREG/CR

-7203 (NRC, 2015) should not be considered 19 generically applicable with respect to criticality safety analyses. 20 Criticality is not a concern for dry SNF systems, as SNF requires moderation to reach criticality. 21 Although DSS casks are expected to remain dry while in storage, cask users may be allowed to 22 load and unload a cask in a wet environment. The criticality analyses in NUREG/CR

-7203 are 23 performed with an assumption of fully flooded conditions and any conclusions adopted are 24 applicable to analyses that support wet loading and unloading. The following considerations for 25 criticality evaluations for reconfigured fuel are applicable only to DSS scenarios where there 26 may be flooding within the canister. Otherwise, the staff does not find reconfiguration to pose a 27 criticality safety concern for a dry system.

28 All of the criticality safety analyses presented in NUREG/CR

-7203 take credit for burned fuel 29 nuclides (burnup credit) and the conclusions may not be applicable to criticality analyses that 30 assume a fresh fuel composition. In its review of the burnup credit methodology and code 31 benchmarking used to support a criticality safety evaluation, the staff will follow the guidance in 32 ISG-8, Revision 3, "Burnup Credit in the Criticality Safety Analyses of PWR Spent Fuel in 33 Transportation and Storage Casks," issued in September 2012 (NRC, 2012) to review the 34 burnup credit analys e s. ISG-8, Revision 3, does not endorse any particular methodology for 35 BWR fuel burnup credit. The staff does not necessarily endorse the methodology described in 36 NUREG/CR-7203 for BWR fuel DSS, and considers it to be for illustration only.

37 For criticality safety analyses using burnup credit, NUREG/CR

-7203 (NRC, 2015) shows that 38 reactivity increases for longer decay times (e.g., analyses supporting storage beyond 20 years

) 39 would need to use an appropriate decay time within the criticality evaluations.

The enrichment 40 and burnup values assumed within the criticality evaluations in NUREG/CR

-7203 may differ 41 from those allowed within another storage system. However NUREG/CR

-7203 states that no 42 significant differences were observed in trends between configurations that evaluated fuel at 43 44.25 GWd/MTU and 70 GWd/MTU.

44 3-16 The following sections discuss information from NUREG/CR-7203 that may be applicable when 1 performing reconfiguration analyses within a criticality evaluation for HBU fuel under normal, off

-2 normal , and accident conditions of storage.

3 Normal Conditions of Storage 4 In an approach acceptable to the staff, t he applicant's criticality safety analys e s would consider 5 the reactivity impact of 3-percent fuel failure during normal conditions of storage. The most 6 applicable scenario from NUREG/CR

-7203 (NRC, 2015) is Scenario 1(a)

(See Section 3.2.4.2 7 above for a description of the scenarios

). 8 ORNL created Scenario 1(a) to represent breached rods. ORNL assumed that a percentage of 9 the rods were breached and that cladding from these rods failed completely and then removed 10 this percentage of fuel rods from the system.

This is conservative as SNF systems are 11 undermoderated and replacing fuel with moderator typically causes reactivity to increase. Using 12 a fresh fuel composition for PWR fuel, ORNL's models in NUREG/CR

-7203 showed that 13 reactivity decreases when removing r ods. Therefore, this type of analysis may not be 14 appropriate for PWR analyses that assume a fresh fuel composition. The location assumed for 15 failed or removed rods can significant ly effect reactivity. ORNL showed in Section A.1.1 of 16 NUREG/CR-7203 that removing rods from the center of the assembly causes reactivity to 17 increase the most.

18 In NUREG/CR-7203, ORNL also showed the number of rods removed that produces the 19 maximum reactivity. For the systems studied , NUREG/CR-7203 shows that the maximum 20 reactivity occurs when a number of rods far greater than 1

-percent is removed from the system.

21 NUREG/CR-7203 also presents the results of a sensitivity study showing that reactivity increases 22 even more for Scenario 1(a) when it is assumed that the failed fuel relocates to a location outside 23 of the absorber plate. This is based on the generic system s modeled for the study. A different 24 system may allow relocation of the failed rod material outside of the absorber plate material to a 25 different extent.

26 Off-Normal Conditions of Storage 27 In an approach acceptable to the staff, t he applicant's criticality safety analys e s would consider 28 the reactivity impact of 10-percent fuel failure under off

-normal conditions of storage. The 29 methods discussed in the previous section on normal conditions of storage also apply to off

-30 normal conditions of storage

however the applicant would consider fuel failure up to 10 percent 31 rather than 1 percent. Scenario 1(a) can be used to represent rod failure via removing rods 32 from the system.

In this case an applicant would remove 10-percent of the rods rather than 1-33 percent. The applicant would remove rods in such a way that it produces maximum reactivity 34 and consider relocation of the fuel to outside of the absorber plates.

35 Accident Conditions of Storage 36 In an approach acceptable to the staff, t he applicant's criticality safety analys e s would consider 37 the reactivity impact of 100-percent fuel failure under accident conditions of storage. The 38 damaged fuel models in Section A.1.2 for Scenario 1(b) from NUREG/CR

-7203 are applicable 39 when representing 100 percent failed fuel.

40 3-17 Scenario 1(b) from NUREG/CR

-7203 considers reconfiguration of damaged fuel. With 100-1 percent compromise in cladding integrity, reconfiguration is considered to the maximum extent.

2 Section A.1.2 of NUREG/CR

-7203 shows that a model assuming an "ordered pellet array" is 3 more reactive than a homogenous mixture of fuel, cladding materials and water.

4 3.2.4.2.5 Shielding 5 An application may demonstrate that a DS S continues to meet the regulatory dose limits for the 6 period beyond 20 years by assuming hypothetical reconfiguration of the HBU SNF into a 7 justified bounding geometric form under normal, off

-normal, and accident conditions. This 8 method is one way to demonstrate compliance with 10 CFR 72.104, 10 CFR 72.106, or 10 CFR 9 72.236(d). 10 To assess the impacts of various fuel geometry changes on the shielding designs of DSSs and 11 ISFSI s, ORNL analyzed various scenarios of fuel geometry changes and the impact on the 12 annual dose at the ISFSI boundary and dose rates near the cask and presented the results in 13 NUREG/CR-7203 (NRC, 2015)

. 14 Appendix B to NUREG/CR-7203 (provides some insight into the effects on external dose for 15 various reconfiguration scenarios

however the results in NUREG/CR

-7203 should not be 16 considered generically applicable with respect to external dose and dose rate evaluations. A 17 DSS designer would assess the impacts of fuel reconfiguration on external dose and dose rates 18 for its particular design. 19 This section discusses an approach acceptable to the staff for addressing the impacts on 20 external dose and dose rates when considering possible reconfiguration of HBU fuel for a period 21 of storage beyond 20 years. This discusses the scenarios from NUREG/CR

-7203 most 22 applicable to the reconfiguration under normal, off

-normal , and accident conditions of storage as 23 well as the analytical assumptions likely to result in bounding dose and dose rates based on the 24 results from NUREG/CR

-7203. The NUREG has considered burnup up to 65 GWd/MTU within 25 its dose and dose rate evaluations. As discussed in Section B.5 of NUREG/CR

-7203, different 26 nuclides become important to external dose and dose rate based on the decay time.

27 Since reconfiguration is to be considered after 20 years of storage, and this length of cooling 28 time is generally much longer than cooling times used to establish loading tables, applicants 29 may be able to make the justification that increases to external dose due to reconfiguration are 30 bounded by the additional cooling time the assemblies will experience.

31 NUREG/CR-7203 also indicates that fuel assembly type, (i.e., PWR vs BWR

), may have a 32 significant impact on the surface dose rate and controlled area boundary dose under fuel 33 reconfiguration scenarios. Table s 13 and 14 of NUREG/CR

-7203 show the difference in dose 34 rate increase for BWR and PWR SNF. A DSS system may permit storage of other fuel 35 assemblies, with different allowable burnup and enrichments to which the results of 36 NUREG/CR-7203 (NRC, 2015) do not apply. The burnup profile and depletion parameters used 37 to create the source term within NUREG/CR

-7203 may also not be generically applicable.

38 Normal Conditions of Storage 39 In an approach acceptable to the staff, t he applicant's external dose and dose rate evaluation 40 would consider the impact of 3-percent fuel failure during normal conditions of storage. The 41 most applicable scenario from NUREG/CR

-7203 is Category 1, fuel failure, Scenario, 1(a). If 42 3-18 cladding is breached and the fuel fails, this could lead to source relocation or change of the 1 geometric shape of the source. Based on NUREG/CR

-7203, the impact on the controlled

-area 2 boundary dose caused by source relocation resulting from 1-percent fuel failure is insignificant.

3 For a different DSS, the application may need to discuss potential fuel failure and source 4 reconfiguration and the potential impact on controlled

-area boundary doses as required by 10 5 CFR 72.104 and 10 CFR 72.106. 6 Depending on the DSS and the resultant fuel geometry, the dose rate may increase significantly 7 as the detector moves close to the cask. Although it may not cause a significant change to the 8 dose and therefore may not constitute a significant concern for people at the controlled area 9 boundary, the changes of source term geometry wi ll affect the doses of occupational workers 10 who need to perform necessary work around the casks.

In general, an application should 11 consider the impact of HBF failure on the near cask dose rate and potential impacts on radiation 12 protection associated with ISFSI surveillance and maintenance operations.

13 Off-Normal Conditions of Storage 14 In an approach acceptable to the staff, t he applicant's external dose and dose rate evaluation 15 for HBF would consider the impact of 10-percent fuel failure under off-normal conditions of 16 storage. If cladding is breached and fails, the fuel, and hence the source, may relocate to 17 different parts of the fuel basket. The impact of HBFfailure on dose at the controlled

-area 18 boundary for storage under off

-normal conditions of dry storage operations should be examined.

19 A 10-percent fuel failure is similar to Scenario 1(a) in NUREG/CR

-7203 (NRC, 2015). For 20 Scenario 1(a), breached rods, ORNL assumed the rods turned to rubble and calculated the 21 dose rate when the fuel mixture relocated to the bottom of the fuel assembly. ORNL assumed 22 failure of 10-percent of fuel rods collected into the available free volume within the assembly 23 lower hardware region. Section B.4.1 of NUREG/CR

-7203 discusses the implementation in 24 detail. ORNL reduced the source strength and density of the active fuel zone by the failure 25 percentage and relocated this source to the bottom of the fuel assembly and increased the 26 source strength and density accordingly. The storage system in NUREG/CR

-7203 is modeled 27 as a vertically-oriented storage system. Fuel would likely not relocate this way in a horizontal 28 storage system

, and the models is not necessarily applicable to a horizontal system.

29 In Section B.5.5 of NUREG/CR

-7203 , ORNL discuss the results of the study performed on the 30 individual storage, which shows that there could be significant increases in the dose rate near 31 the cask. It concludes that fuel configuration changes can cause significant dose rate increases 32 relative to the nominal intact fuel configuration in the cask outer regions that face air vent 33 locations. NUREG/CR-7203 states that the change in radiation dose rate a way from air vent 34 locations is either small or negligible.

35 Similar to normal conditions of storage, the changes in source term geometry will impact the 36 doses of occupational workers who need to perform necessary surveillance and maintenance 37 work around the casks. To assess the impacts on radiation protection, an applicant may need 38 to evaluate the surface dose rate increase resulting from reconfiguration.

39 Accident Conditions of Storage 40 In an approach acceptable to the staff, t he applicant's external dose and dose rate evaluation 41 for HBF would consider the impact of 100-percent fuel failure during normal conditions of 42 storage. If cladding is breached and the fuel fails, this may cause the fuel, and hence the 43 3-19 source, to relocate to different parts of the fuel basket. Based NUREG/CR

-7203 (NRC, 2015), 1 the impacts on the controlled

-area boundary dose caused by source relocation resulting from 2 100 percent fuel failure w ill result in significant increases in the dose rate near the cask and 3 annual dose on the controlled area boundary

. Scenarios 1(b) and 2 in NUREG/CR 7203 can 4 represent 100-percent fuel failure. 5 At the controlled area boundary, 100-percent fuel reconfiguration can have a significant impact 6 on the annual dose. It can also significantly affect the dose rate near the cask and the radiation 7 protection associated with ISFSI remediation operations. Table s B.9 and B.10 of Appendix B t o 8 NUREG/CR-7203 (NRC, 2015) show the relative changes in dose rates at 1 meter from a 9 sample PWR fuel cask and a sample BWR fuel cask, respectively. Table B.11 of Appendix B to 10 NUREG/CR-7203 shows the estimated relative impact on controlled

-area boundary dose from 11 fuel reconfiguration. The data presented in these tables show that the impacts on the dose 12 rates at the cask side, particularly the dose rate near the vent ports are significant.

13 In Scenario 1(b), ORNL assumed that the assembly and basket plate material is homogenized , 14 placed it at the bottom of the cask

, and determined that the limiting packing fraction is 0.58.

15 This scenario did not produce an increase in site boundary dose

however, it did show an 16 increase in local dose rates. The location of the "bottom" of the cask would depend on whether 17 the DSS is vertical or horizontal. Homogenizing the basket material with the fuel rubble may be 18 overly conservative for a horizontal configuration, and applicants may choose to maintain basket 19 integrity similar to the Scenario S2 model in Section B.4.2 of NUREG/CR-7203 when evaluating 20 dose or dose rates for a horizontal system or a tip-over scenario.

21 For Scenario 1(b), ORNL also assumed that the fuel and basket material forms a homogenized 22 rubble that is distributed throughout the canister cavity. This scenario produced an increase in 23 site boundary dose.

24 3.3 Canned Fuel (Damaged Fuel) 25 10 CFR 72.122(h)(1) requires SNF, including HBU , with gross ruptures (i.e., classified as 26 damaged) be placed in a can designed for damaged fuel or in an acceptable alternative.

The 27 staff will follow the guidance in the current SRPs for dry storage of SNF in its review of an 28 application for a DSS with damaged HBU SNF contents

. 29