Clad Degradation – FEPs Screening Arguments Rev 00, ICN 00 ANL-WIS-MD-000008 April 2000 1. PURPOSE The purpose of this report is to summarize the work of others into the Features, Events, and Processes (FEPs) for cladding degradation screening that is consistent with and used in the Total System Performance Assessment - Site Recommendation (TSPA-SR). This review is to describe what FEPs are to be included in the TSPA-SR and also document the reason for excluding others. In accordance with AP-2.13Q, Technical Product Development Planning, a work plan (CRWMS M&O 1999a) was developed, issued, and utilized in the preparation of this document. Under the provisions of the U.S. Department of Energy’s (DOE’s) Revised Interim Guidance Pending Issuance of New U. S. Nuclear Regulatory Commission (NRC) Regulations (Revision 01, July 22, 1999), for Yucca Mountain Nevada (Dyer 1999; and herein referred to as DOE’s interim guidance), the DOE must provide a reasonable assurance that the regulatory-specified performance objectives for the Yucca Mountain Project can be achieved for a 10,000-year postclosure period. This assurance must be demonstrated in the form of a performance assessment that: (1) identifies the features, events, and processes (FEPs) that potentially affect the performance of the geologic repository; (2) examines the effects of such FEPs on the performance of the geologic repository; (3) estimates the expected annual dose to a specified nearby population group. The performance assessment must also provide the technical basis for inclusion or exclusion of specific FEPs. Although the NRC has not defined nor used the term "scenario" in the pertinent regulations, the Yucca Mountain Total System Performance Assessment - Site Recommendation(TSPA-SR) has chosen to satisfy the above-stated performance assessment requirements by adopting a scenario development process. This decision was made based on the Yucca Mountain TSPA-SR adopting a definition of "scenario" as not being limited to a single, deterministic feature of the system, but rather as a set of similar features that share common FEPs. The DOE has chosen to adopt a scenario development process based on the methodology developed by Cranwell et al. (1990) for the NRC. The first step of this process is the identification of FEPs potentially relevant to the performance of the Yucca Mountain repository; the second step includes the screening of each FEP. The primary purpose of this Analysis/Model Report (AMR) is to identify and document the analysis, screening decision, and Total System Performance Assessment (TSPA-SR) disposition or screening argument for FEPs related to clad degradation. Both primary and secondary FEPs are considered. The screening decisions, and associated TSPA-SR disposition or screening argument, for the subject FEPs will be cataloged separately in a project-specific FEPs database (see Section 1.4). This AMR and the database are being used to document information related to the FEPs screening decisions and associated screening argument and to assist reviewers during the license review process. Most commercial nuclear fuel is encased in Zircaloy cladding. This AMR does not address potential damage to assemblies that might occur at the YMP surface facilities. There are constraints, caveats and limitations to this report. This FEPs screening is based on commercial Pressurized Water Reactor (PWR) fuel with Zircaloy cladding but is applicable to Boiling Water Reactor (BWR) fuel. This analysis referenced in the other AMRs is also limited to fuel exposed to normal operation and anticipated operational occurrences (i.e. events which are anticipated to occur within a reactor lifetime), and is not applicable to fuel that has been exposed to severe accidents. Fuel burnup projections have been limited to the current commercial reactor licensing environment with restrictions on fuel enrichment, oxide coating thickness, and rod plenum pressures. Ranges and uncertainties have been defined in the other AMRs. The information provided in this FEPs screening will be used in evaluating what cladding degradation mechanisms are included in the post-closure performance of the Monitored Geologic Repository (MGR) in relation to waste form degradation. 1.1 SCOPE This AMR has been prepared to satisfy the FEP screening documentation requirements in the Work Scope/Objectives/Tasks section of the development plan entitled Cladding FEPs Screening Arguments (CRWMS M&O 1999a). The current FEPs list consists of 1786 entries (as described in Section 1.2). The FEPs have been classified as primary and secondary FEPs (as described in Section 1.2) and have been assigned to various Process Modeling Reports (PMRs). The assignments were based on the nature of the FEPs so that the analysis and resolution for screening decisions reside with the subject-matter experts in the relevant disciplines. The resolution of other than EBS FEPs is documented in AMRs prepared by the responsible PMR groups. Several relevant FEPs do not fit into the existing PMR structure. Criticality is the largest example, and is treated in FEP assignments as if were a separate PMR. Some FEPs were best assigned to the TSPA-SR itself (i.e., system-level FEPs), rather than to its component models. This AMR addresses the clad degradation FEPs. These FEPs represent the key features that result in degradation of the cladding. The clad degradation primary and secondary EBS FEPs addressed in this AMR are provided in Table 1. ANL-WIS-MD-000008 REV 00 9 April 2000 Table 1. FEPs Related to the CSNF Clad Degradation Component FEP Number FEP Title 2.1.02.07.00 Gap and Grain Release of Cs, I 2.1.02.11.00 Waterlogged Rods 2.1.02.12.00 Cladding Degradation Before YMP Receives It 2.1.02.12.01 Pin Degradation During Reactor Operation 2.1.02.12.02 Pin Degradation During Spent Fuel Pool Storage 2.1.02.12.03 Pin Degradation During Dry Storage 2.1.02.12.04 Pin Degradation During Fuel Shipment and Handling 2.1.02.13.00 General Corrosion of Cladding 2.1.02.13.01 Cladding Degradation Mechanisms At YMP, Pre-Pin Failure 2.1.02.13.02 Corrosion (of Cladding) 2.1.02.14.00 Microbiologically Influenced Corrosion (MIC) of Cladding 2.1.02.15.00 Acid Corrosion of Cladding From Radiolysis 2.1.02.16.00 Localized (Pitting) Corrosion of Cladding 2.1.02.17.00 Localized Corrosion (Crevice Corrosion) of Cladding 2.1.02.18.00 High Dissolved Silica Content of Waters Enhances Corrosion of Cladding 2.1.02.19.00 Creep Rupture of Cladding 2.1.02.20.00 Pressurization From He Production Causes Cladding Failure 2.1.02.21.00 Stress Corrosion Cracking (SCC) of Cladding 2.1.02.21.01 Inside Out From Fission Products (Iodine) (Failure of Cladding) 2.1.02.21.02 Outside In From Salts Or WP Chemicals (Failure of Cladding) 2.1.02.22.00 Hydride Embrittlement of Cladding 2.1.02.22.01 Hydride Embrittlement From Zirconium Corrosion (of Cladding) 2.1.02.22.02 Hydride Embrittlement From WP Corrosion & H2 Absorption (of Cladding) 2.1.02.22.03 Hydride Embrittlement From Galvanic Corrosion of WP Contacting Cladding 2.1.02.22.04 Delayed Hydride Cracking (of Cladding) 2.1.02.22.05 Hydride Reorientation (of Cladding) 2.1.02.22.06 Hydrogen Axial Migration (of Cladding) 2.1.02.22.07 Hydride Embrittlement From Fuel Reaction (Causes Failure of Cladding) 2.1.02.23.00 Cladding Unzipping 2.1.02.23.02 Dry Oxidation of Fuel (Causes Failure of Cladding) 2.1.02.23.03 Wet Oxidation of Fuel (Causes Failure of Cladding) 2.1.02.24.00 Mechanical Failure of Cladding 2.1.02.27.00 Localized Corrosion Perforation from Fluoride 2.1.02.28.00 Diffusion-Controlled Cavity Growth 2.1.7.01.00 Rockfall (Large Block) 2.1.09.03.00 Volume Increase of Corrosion Products 2.1.11.07.00 Thermally-Induced Stress Changes in Waste and EBS 1.2 FEPs IDENTIFICATION AND ANALYSIS For the YMP TSPA-SR, a scenario is defined as a subset of the set of all possible features of the fuel cladding that contains the features resulting from a specific combination of FEPs. The first step of the scenario development process is the identification of FEPs potentially relevant to the performance of the Yucca Mountain repository. The initial set of FEPs was created for the Yucca Mountain TSPA-SR by combining lists of FEPs previously identified as relevant to the Yucca Mountain Project (YMP) (e.g. Total System Performance Assessment-1995: An Evaluation of the Potential Yucca Mountain Repository, CRWMS M&O 1995a) with a draft FEP list compiled by the Nuclear Energy Agency (NEA) of ANL-WIS-MD-000008 REV 00 10 April 2000 the Organization for Economic Cooperation and Development (OECD). The NEA list is maintained as an electronic FEP database and is the most comprehensive list available internationally. The list currently contains 1261 FEPs from Canadian, Swiss, and Swedish spentfuel programs, intermediate and low-level waste programs of the U.K., and the U.S. Waste Isolation Pilot Plant (WIPP) program. An additional 292 FEPs have been identified from YMP literature and site studies, and 82 FEPs have been identified during YMP project staff workshops. These FEPs are organized under 151 categories, based on NEA category headings, resulting in a total of 1786 entries. Consistent with the diverse backgrounds of the programs contributing FEPs lists, FEPs have been identified by a variety of methods, including expert judgment, informal elicitation, event tree analysis, stakeholder review, and regulatory stipulation. All potentially relevant FEPs have been included, regardless of origin. This approach has led to considerable redundancy in the FEP list, because the same FEPs are frequently identified by multiple sources, but it also ensures that a comprehensive review of narrowly defined FEPs will be performed. The FEPs list is considered open and will continue to grow as additional FEPs are identified. There is no uniquely correct level of detail at which to define scenarios or FEPs. Decisions regarding the appropriate level of resolution for the analysis are made based on consideration of the importance of the scenario in its effect on overall performance and the resolution desired in the results. The number and breadth of scenarios depend on the resolution at which the FEPs have been defined: coarsely defined FEPs result in fewer, broad scenarios, whereas narrowly defined FEPs result in many narrow scenarios. For efficiency, both FEPs and scenarios should be aggregated at the coarsest level at which a technically sound argument can be made that is adequate for the purposes of the analysis. Consequently, each FEP has been identified as either a primary or secondary FEP. Primary FEPs are those FEPs for which the project proposes to develop detailed screening arguments. The classification and description of primary FEPs strive to capture the essence of all the secondary FEPs that map to the primary. For example, the primary FEP "Cladding Degradation Before YMP Receives It" can be used appropriately to resolve multiple and redundant secondary FEPs that address degradation during reactor operation, during spent fuel pool storage, etc. By working to the primary FEP description, the subject matter experts assigned to the primary FEP address all relevant secondary FEPs, and arguments for secondary FEPs can be rolled into the primary FEP analysis. Secondary FEPs are FEPs either that are completely redundant or that can be aggregated into a single primary FEP. To perform the screening and analysis, the FEPs have been assigned so that the analysis, screening decision, and TSPA-SR disposition reside with the subject matter experts in the relevant disciplines. The TSPA-SR recognizes that FEPs have the potential to affect multiple facets of the project, may be relevant to more than one PMR, or may not fit neatly within the PMR structure. For example, many FEPs affect waste form (WF), waste package (WP), and the EBS. Rather than create multiple separate FEPs, the FEPs have been assigned, as applicable, to one or more process modeling groups, which are responsible for the AMRs. At least two approaches have been used to resolve overlap and interface problems of multiple numbers for the same FEP. FEP owners from different process modeling groups may decide that only one PMR will address all aspects of the FEP, including those relevant to other PMRs. ANL-WIS-MD-000008 REV 00 11 April 2000 Alternatively, FEP owners may each address only those aspects of the FEP relevant to their area. In either case, the FEP AMR produced by each process modeling group lists the FEP and summarizes the screening result, citing the appropriate work in related AMRs as needed. 1.3 FEPs SCREENING AND ANALYSIS PROCESS As described in Section 1.2, the first step in the scenario development process was the identification and analysis of FEPs. The second step in the scenario development process includes the screening of each FEP. Each FEP is screened for inclusion or exclusion in the TSPA-SR against three criteria, which are stated as regulatory requirements in NRC’s proposed rule 10 CFR Part 63 (64 FR 8640), and in the U.S. Environmental Protection Agency's (EPA) proposed rule 40 CFR Part 197 (64 FR 46976). The screening criteria are discussed in more detail in Section 4.2 and are summarized here. FEPs are excluded from the TSPA-SR only if: · They are specifically ruled out by regulation, are contrary to the stated regulatory assumptions, or are in conflict with statements made in background information regarding intent or directions of the regulations. · They can be shown to have a probability of occurrence of less than 10-4 in 104 years. · Their occurrence can be shown to have no significant effect on the overall TSPA. The regulatory screening criteria contained in DOE’s interim guidance (Dyer 1999) and in the proposed 40 CFR Part 197 (64 FR 46976) are relevant to many of the FEPs. FEPs that are contrary to DOE’s interim guidance or to specific proposed regulations, regulatory assumptions, or regulatory intent are excluded from further consideration. Examples include: the explicit exclusion of all but a specified scenario to address treatment of human intrusion (10 CFR §63.113(d)), assumptions about the critical group to be considered in the dose assessment (10 CFR §63.115), and the intent that the consideration of "the human intruders" be excluded from the human intrusion assessment (64 FR 8640, Section XI. Human Intrusion). Probability estimates used in the FEPs screening process may be based on technical analysis of the past frequency of similar events (such as igneous and seismic events) or, in some cases, on expert elicitation. Probability arguments, in general, require including some information about the magnitude of the event in its definition. Probability arguments are also sensitive to the spatial and temporal scales at which FEPs are defined. For example, the definition of the probability of a seismic event depends on the magnitude of the event. Probability arguments are therefore made at reasonably coarse scales. Consequence-based screening arguments can be established in a variety of ways. Various methods include TSPA-SR sensitivity analyses, modeling studies outside of the TSPA-SR, or reasoned arguments based on literature research. For example, consequences of many geomorphic processes such as erosion and sedimentation can be evaluated by considering bounding rates reported in geologic literature. More complicated processes, such as igneous activity, require detailed analyses conducted specifically for the Yucca Mountain Project. Lowconsequence arguments are often made by demonstrating that a particular FEP has no effect on the distribution of an intermediate performance measure in the TSPA-SR. For example, by ANL-WIS-MD-000008 REV 00 12 April 2000 demonstrating that including a particular waste form has no effect on the concentrations of radionuclides transported from the repository in the aqueous phase, it is also demonstrated that including this waste form in the inventory would not compromise compliance with the performance objectives. Explicit modeling of the characteristics of this waste form could therefore be excluded from the TSPA-SR. Using the type of arguments discussed above, each FEP identified as relevant to the clad degradation was reviewed against the three exclusion criteria. Those that were determined to meet one of the three criteria were designated as “excluded” from further consideration within the TSPA-SR. Those that did not meet any of these criteria must, by definition, be “included.” 1.4 ORGANIZATION OF FEP DATABASE Under a separate scope, the TSPA-SR team is constructing an electronic FEP database to assist project reviewers during the license review process. Each FEP has been entered as a separate record in the database. Fields within each record provide a unique identification number, a description of the FEP, the origin of the FEP, identification as a primary or secondary FEP for the purposes of the TSPA-SR, and mapping to related FEPs and to the assigned PMRs. Fields also provide summaries of the screening arguments with references to supporting documentation and AMRs, and, for all included or retained FEPs, statements of the disposition of the FEP within the TSPA-SR modeling system. The AMRs, however, contain the detailed arguments and description of the disposition of the subject FEPs. Alphanumeric identifiers (called the “NEA category”) previously used have been retained in the database for traceability purposes. Each FEP has also been assigned a unique YMP FEP database number, based on the NEA categories. The database number is the primary method for identifying FEPs, and consists of an eight-digit number of the form x.y.zz.pp.qq. The general structure of the database is reflected in the first two digits (x.y) as shown below: 0.0. Assessment Basis 1.0. External Factors 1.1 Repository Issues 1.2 Geological Processes and Effects 1.3 Climatic Processes and Effects 1.4 Feature Human Actions (Active) 1.5 Other 2.0. Disposal System - Environmental Factors 2.1 Wastes and Engineered Features 2.2 Geologic Environment 2.3 Surface Environment 2.4 Human Behavior 3.0. Disposal System - Radionuclide/Contaminant Factors 3.1 Contaminant Characteristics 3.2 Contaminant Release/Migration Factors 3.3 Exposure Factors ANL-WIS-MD-000008 REV 00 13 April 2000 The next six digits (zz.pp.qq) define a grouping structure for the FEPs, with zz designating the category, and pp designating the heading. The exact details of this grouping structure are not important to the evaluation, since each FEP will be evaluated regardless of the database organization. Finally, the last two digits (qq) signify whether the FEP is primary (00) or secondary (other than 00). Each heading has a primary FEP associated with it, and may or may not have any secondary FEPs. In those cases where secondary FEPs do exist, the primary FEP encompasses all the issues associated with the secondary FEPs. The secondary FEPs either provide additional detail concerning the primary, or are a restatement of the primary based on redundant input from a different source. 2. QUALITY ASSURANCE The activities documented in this AMR were evaluated in accordance with QAP-2-0, Conduct of Activities and were determined to be quality affecting and subject to the requirements of the U.S. DOE Office of Civilian Radioactive Waste Management (OCRWM) Quality Assurance Requirements and Description (QARD) (DOE 2000). This evaluation is documented in Conduct of Performance Assessment (CRWMS M&O 1999c). Accordingly, analysis activities documented in this AMR have been conducted in accordance with the Civilian Radioactive Waste Management System Management and Operating Contractor (CRWMS M&O) quality assurance program, using approved procedures identified in Cladding FEPs Screening Arguments (CRWMS M&O 1999a). This AMR has been developed in accordance with procedure AP-3.10Q, Analyses and Models. Preparation of this analysis did not require the classification of items in accordance with QAP-2- 3, Classification of Permanent Items. This activity is not a field activity. Therefore, an evaluation in accordance with NLP-2-0, Determination of Importance Evaluations was not required. 3. COMPUTER SOFTWARE AND MODEL USAGE This AMR uses no computational software nor model. The AMR was documented using only commercially available software (Microsoft Word 97-SR2) for word processing, which is exempt from qualification requirements in accordance with AP-SI.1Q, Software Management. There were no additional applications (routines or macros) developed using this commercial software. The analyses and arguments presented herein are based on regulatory requirements, results of analyses presented and documented in other AMRs, or technical literature. 4. INPUTS There are no input data sources used in this analysis. 4.1 Data and Parameters This section is not applicable to this analysis. ANL-WIS-MD-000008 REV 00 14 April 2000 4.2 Criteria Programmatic requirements for this document are listed in the development plan (CRWMS M&O 1999a), which covers eight tasks needed to provide the submodels for the performance assessment Physical and Chemical Environment Abstraction Model. The development plan specifies that this document and all analyses described herein must adhere to the requirements of AP-3.10Q, Analyses and Models, and must address applicable NRC issue resolution status report (IRSR) criteria for the TSPA (NRC 1998). The U.S. Nuclear Regulatory Commission’s Total System Performance Assessment and Integration (TSPAI) Issue Resolution Status Report (NRC 1998) establishes generic technical acceptance criteria considered by the NRC staff to be essential to a defensible, transparent, and comprehensive assessment methodology for the repository system. The NRC’s IRSR Key Technical Issue: Container Life and Source Term (NRC 1999) establishes generic technical acceptance criteria considered by the NRC staff for the waste form, with the cladding degradation analysis being part of this KTI. Attachment IV of Initial Cladding Condition (CRWMS M&O 2000b) describes how IRSR issues and criteria are addressed. Technical screening criteria are provided in DOE’s interim guidance (Dyer 1999) and have also been identified by the NRC in the proposed 10 CFR Part 63 (64 FR 8640) and by the EPA in the proposed 40 CFR Part 197 (64 FR 46976). Both proposed regulations specifically allow the exclusion of FEPs from the TSPA-SR if they are of low probability (less than one chance in 10,000 of occurring in 10,000 years) or if occurrence of the FEP can be shown to have no significant effect on expected annual dose. There is no quantified definition of “significant effect” in the guidance nor proposed regulations. 4.2.1 Low Probability The probability criterion is explicitly stated by the NRC in the proposed 10 CFR §63.114 (d): Consider only events that have at least one chance in 10,000 of occurring over 10,000 years. The EPA provides essentially the same criterion in 40 CFR §197.40: The DOE’s performance assessments should not include consideration of processes nor events that are estimated to have less than one chance in 10,000 of occurring within 10,000 years of disposal. 4.2.2 Low Consequence Criteria for low consequence screening arguments are provided in DOE’s interim guidance (Dyer 1999, Section 114(e) and (f)), which indicates that performance assessment shall: (e) Provide the technical basis for either inclusion or exclusion of specific features, events, and processes of the geologic setting in the performance assessment. Specific features, events, and processes of the geologic setting must be evaluated in ANL-WIS-MD-000008 REV 00 15 April 2000 detail if the magnitude and time of the resulting expected annual dose would be significantly changed by their omission. (f) Provide the technical basis for either inclusion or exclusion of degradation, deterioration, or alteration processes of engineered barriers in the performance assessment, including those processes that would adversely affect the performance of natural barriers. Degradation, deterioration, or alteration processes of engineered barriers must be evaluated in detail if the magnitude and time of the resulting expected annual dose would be significantly changed by their omission. The EPA provides essentially the same criteria in 40 CFR §197.40: …with the NRC’s approval, the DOE’s performance assessment need not evaluate, in detail, the impacts resulting from any processes and events or sequences of processes and events with a higher chance of occurrence if the results of the performance assessment would not be changed significantly. The terms “[not] significantly changed” and “not .. changed significantly” are undefined terms in DOE’s interim guidance and in the EPA’s proposed regulations. These terms are inferred for FEPs screening purposes to be equivalent to having no or negligible effect. Because the relevant performance measures differ for different FEPs (e.g., effects on performance can be measured in terms of changes in concentrations, flow rates, travel times, and other measures, as well as overall expected annual dose), there is no single quantitative test of “significance.” 4.2.3 Reference Biosphere Both DOE’s interim guidance (Dyer 1999) and EPA’s proposed regulations specify assumptions (which in effect serve as criteria) pertinent to screening many of the EBS FEPs. Particularly germane are explicit assumptions regarding the reference biosphere (10 CFR §63.115), and less so are assumptions regarding the location and use of groundwater by the critical group used for calculation of exposure doses. The assumptions pertaining to the characteristics of the reference biosphere are presented in DOE’s interim guidance (Dyer 1999, Section 115(a)1). The specified characteristics pertinent to the EBS FEPs are that: · Features, events, and processes …shall be consistent with present knowledge of the conditions in the region surrounding the Yucca Mountain site. The EPA has specified a similar assumption in proposed 40 CFR §197.15. This assumption is stated as: … DOE must vary factors related to the geology, hydrology, and climate based on environmentally protective but reasonable scientific predictions of the changes that could affect the Yucca Mountain disposal system over the next 10,000 years. ANL-WIS-MD-000008 REV 00 16 April 2000 4.3 Codes and Standards ASTM C1174-97, Standard Practice for the Long-Term Behavior of Materials, Including Waste Forms, Used in Engineered Barrier Systems (EBS) for Geologic Disposal of High-Level Radioactive Waste is used to support the degradation analysis development methodology, categorize the analyses developed with respect to their usage for long-term TSPA-SR, and relate the information/data used to develop the analysis to the requirements of the standard. Material standards are also cited. ASTM B 551 and ASTM B811 are referenced in Section 6.2.12. 5. ASSUMPTIONS This AMR makes no assumptions beyond those reported and documented by the references. 6. ANALYSIS/MODEL - FEPs RELATED TO CLADDING DEGRADATION To demonstrate that the regulatory-specified performance objectives of proposed 10 CFR Part 63 and proposed 40 CFR Part 197 can be achieved for a 10,000-year post-closure period, the Yucca Mountain Project is implementing a stochastic scenario development methodology based on the work of Cranwell et al. (1990). The methodology provides a systematic approach for considering, as completely as practicable, the possible feature states of a repository system. It seeks to span the set of all possible feature states using a finite set of scenarios. Each scenario represents the ensemble of possible feature states corresponding to parameter and model uncertainties present in the group of FEPs composing the scenario. The methodology begins with a comprehensive FEP identification step followed by a rigorous FEP screening step. With its focus on FEPs related to cladding degradation, this Analysis/Modeling Report (AMR) considers these first two steps (i.e., FEP identification and FEP screening) of scenario development. 6.1 Identification of FEPs The first step of the scenario development methodology is the identification of FEPs potentially relevant to the performance of the Yucca Mountain repository. This process is summarized in Section 1.2. Each FEP has been identified as either a “primary” or a “secondary” FEP. Primary FEPs are those FEPs for which the project proposes to develop detailed screening arguments. The classification and description of primary FEPs strive to capture the essence of each secondary FEP that maps to the primary FEP. For example, the primary FEP Meteorite Impact can be used appropriately to resolve multiple and redundant secondary FEPs that address size and effects of meteorite impacts. By working to the primary FEP description, the subject matter experts assigned to the primary FEP will address each relevant secondary FEP, and arguments for secondary FEPs will be rolled into the primary FEP analysis. Table 2 identifies both the primary and the secondary FEPs related to cladding degradation. ANL-WIS-MD-000008 REV 00 17 April 2000 6.2 Screening of FEPs The second step in the scenario development methodology includes the screening of each FEP. Each FEP is screened for inclusion in or exclusion from the TSPA-SR against three criteria, which are stated as regulatory requirements in the NRC proposed rule 10 CFR Part 63 and in EPA proposed rule 40 CFR Part 197. FEPs are excluded from the TSPA-SR for one of the following reasons: (1) They are specifically ruled out by regulation, are contrary to the stated regulatory assumptions, or are in conflict with statements made in background information regarding intent or directions of the regulations. (2) They can be shown to have a probability of occurrence of less than 10-4 in 104 years. (3) Their occurrence can be shown to have no significant effect on the overall TSPA. Examples of reason (1) are the explicit exclusion by the National Academy of Science from consideration of all but a specific scenario to address treatment of human intrusion (10 CFR §63.113(d)) and assumptions about the critical group to be considered in the dose assessment (10 CFR §63.115). Probability estimates used in the FEP screening process in relation to reason (2) may be based on technical analysis of the past frequency of similar events (such as seismic events), or, in some cases, on expert elicitation. Probability arguments, in general, require including appropriate information about the magnitude of the event in its definition. Probability arguments are also sensitive to the spatial and temporal scales at which FEPs are defined. For example, definition of the probability of meteorite impact depends on the size of the meteorite of interest and must consider that meteorite impacts are less likely in shorter time intervals and at smaller locations. Probability arguments are therefore made at reasonably coarse scales based on the judgment of the originator. Consequence based screening arguments in relation to reason (3) can be established in a variety of ways. Various methods include TSPA-SR sensitivity analyses, modeling studies outside of the TSPA-SR, or reasoned arguments based on literature research. For example, consequences of many geomorphic processes such as erosion and sedimentation can be evaluated by considering bounding rates reported in geologic literature. More complicated processes, such as criticality, require detailed analyses conducted specifically for the Yucca Mountain Project. Low consequence arguments are often made by demonstrating that a particular FEP has no effect on the distribution of an intermediate performance measure of the TSPA-SR. For example, to demonstrate that including a particular waste form does not compromise compliance with performance objectives, one may show that aqueous phase concentrations of radionuclides transported from the repository would be unaffected by the exclusion of the specific waste form. Explicit modeling of the characteristics of this waste form could then be excluded from the TSPA-SR. Table 2 presents the “screening decision” as either “include” or “exclude” for FEPs related to cladding degradation. However, for those cases in which there are secondary FEPs, one should be aware of the rule for assigning a composite decision to the primary FEP. Specifically, a primary FEP is designated as include if at least one of its secondary FEPs is designated as ANL-WIS-MD-000008 REV 00 18 April 2000 include. Conversely, a primary FEP is designated as exclude only if all of its secondary FEPs are also designated as exclude. Section 6.2 provides detailed documentation of both the screening argument and the TSPA-SR disposition for each of the primary FEPs. Table 2 lists the specific subsection where each FEP is discussed. If the primary FEP has no secondary FEPs, then the decision to include or exclude the primary FEP is made for that FEP itself. Also provided in each FEP section is a cross reference to key technical issues identified by the NRC as being important for the Yucca Mountain repository. These are identified as Issue Resolution Status Report (IRSR) issues. The key technical issues and subissues are listed below. The relevance of these subissues to the EBS FEPs is identified in Sections 6.2.1 through 6.2.15. Whenever the key technical issue (CLST, for example) is identified rather than a specific subissue, all subissues apply. All the FEPs discussed in this AMP fall into the container life and source term or the Total System Performance Assessment and Integration (TSPAI) subissues which are listed below: Container Life and Source Term (CLST) CLST1 The effects of corrosion processes on the lifetime of the containers CLST2 The effects of phase stability of materials and initial defects on the mechanical failure and lifetime of the containers CLST3 The rate at which radionuclides in spent nuclear fuel (SNF) are released from the EBS through the oxidation and dissolution of spent fuel CLST4 The rate at which radionuclides in high-level waste (HLW) glass are leached and released from the EBS CLST5 The effect of in-package criticality on WP and EBS performance CLST6 The effect of alternate EBS design features on container lifetime and radionuclide release from the EBS Total System Performance Assessment and Integration (TSPAI) TSPAI1 Demonstration of the overall performance objective TSPAI2 Demonstration of multiple barriers TSPAI3 Model abstraction TSPAI4 Scenario analysis TSPAI5 Transparency and traceability of the analysis ANL-WIS-MD-000008 REV 00 19 April 2000 Table 2. Screening Decisions for FEPS Related to Cladding Degradation YMP FEP Database Number NEA Category FEP Name Primary (P) or Secondary (S) FEP Screening Decision Section 2.1.02.07.00 N/A Gap and Grain Release of Cs, I P Include 6.2.13.4 2.1.02.11.00 WF-3 Waterlogged Rods P Exclude 6.2.1 2.1.02.12.00 2.1.03ca Cladding Degradation Before YMP Receives It P Include 6.2.2 2.1.02.12.01 2.1.03cb Pin Degradation During Reactor Operation S Include 6.2.2.1 2.1.02.12.02 2.1.03cc Pin Degradation During Spent Fuel Pool Storage S Exclude 6.2.2.2 2.1.02.12.03 2.1.03cd Pin Degradation During Dry Storage S Include 6.2.2.3 2.1.02.12.04 2.1.03ce Pin Degradation During Fuel Shipment and Handling S Include 6.2.2.4 2.1.02.13.00 2.1.03ch General Corrosion of Cladding P Exclude 6.2.3 2.1.02.14.00 2.1.03ci Microbiologically Influenced Corrosion (MIC) of Cladding P Exclude 6.2.4 2.1.02.15.00 2.1.03cj Acid Corrosion of Cladding From Radiolysis P Exclude 6.2.5 2.1.02.16.00 2.1.03ck Localized Corrosion (Pitting) of Cladding P Exclude 6.2.6 2.1.02.17.00 2.1.03cl Localized Corrosion (Crevice Corrosion) of Cladding P Exclude 6.2.7 2.1.02.18.00 2.1.09f High Dissolved Silica Content of Waters Enhances Corrosion of Cladding P Exclude 6.2.8 2.1.02.19.00 2.1.03cm Creep Rupture of Cladding P Include 6.2.9 2.1.02.20.00 2.1.03cn Pressurization From He Production Causes Cladding Failure P Include 6.2.10 2.1.02.21.00 2.1.03cp Stress Corrosion Cracking (SCC) of Cladding P Include 6.2.11 2.1.02.21.01 2.1.03cq Inside Out From Fission Products (Iodine) (Failure of Cladding) S Include 6.2.11 2.1.02.21.02 2.1.03cr Outside In From Salts Or WP Chemicals (Failure of Cladding) S Include 6.2.11 2.1.02.22.00 2.1.03cs Hydride Embrittlement of Cladding P Exclude 6.2.12 2.1.02.22.01 2.1.03ct Hydride Embrittlement From Zirconium Corrosion (of Cladding) S Exclude 6.2.12.1 2.1.02.22.02 2.1.03cu Hydride Embrittlement From WP Corrosion & H2 Absorption (of Cladding) S Exclude 6.2.12.2 ANL-WIS-MD-000008 REV 00 20 April 2000 YMP FEP Database Number NEA Category FEP Name Primary (P) or Secondary (S) FEP Screening Decision Section 2.1.02.22.03 2.1.03cv Hydride Embrittlement From Galvanic Corrosion of WP Contacting Cladding S Exclude 6.2.12.3 2.1.02.22.04 2.1.03cw Delayed Hydride Cracking (of Cladding) S Exclude 6.2.12.4 2.1.02.22.05 2.1.03cx Hydride Reorientation (of Cladding) S Exclude 6.2.12.5 2.1.02.22.06 2.1.03cy Hydrogen Axial Migration (of Cladding) S Exclude 6.2.12.6 2.1.02.22.07 2.1.03dd Hydride Embrittlement From Fuel Reaction (Causes Failure of Cladding) S Exclude 6.2.12.7 2.1.02.23.00 2.1.03da Cladding Unzipping P Include 6.2.13 2.1.02.23.02 2.1.03db Dry Oxidation of Fuel (Causes Failure of Cladding) S Exclude 6.2.13.1 2.1.02.23.03 2.1.03dc Wet Oxidation of Fuel (Causes Failure of Cladding) S Include 6.2.13.2 2.1.02.24.00 2.1.03co Mechanical Failure of Cladding P Include 6.2.14 2.1.02.27.00 N/A Localized Corrosion Perforation from Fluoride P Include 6.2.15 2.1.02.28.00 N/A Diffusion-Controlled Cavity Growth P Exclude 6.2.16 2.1.07.01.00 N/A Rockfall (Large Block) P Exclude 6.2.14 2.1.09.03.00 N/A Volume Increase of Corrosion Products P Include 6.2.13.3 2.1.11.07.00 N/A Thermally-Induced Stress Changes in Waste and EBS P Include 6.2.9 ANL-WIS-MD-000008 REV 00 21 April 2000 6.2.1 Waterlogged Rods - YMP No. 2.1.02.11.00 FEP Description Failed fuel rods that occur in up to 0.1% of the fuel rods are currently being stored in commercial reactor spent fuel pools. This is attributed to breaches caused by manufacturing defects and reactor operations. Failed fuel contains water in the fuel rod void space. Such fuel is referred to as “waterlogged”. The moisture remaining in a “dried” fuel rod is used to determine the extent of degradation of spent fuel cladding. Screening Decision: Exclude, low consequence. Screening Argument: The fraction of fuel rods with breached cladding is small (up to 0.1%). Moisture can be removed from defective rods during the cask drying operations. The residual moisture of the cask atmosphere can be estimated based on the drying conditions to 5 mbars (= 500 Pa). The impact on the waste form, waste package, and cladding of a small amount of remaining moisture is negligible. TSPA Disposition Few rods are breached at the time they are received at the repository. Breached rods are dried effectively using standard cask drying techniques. The remaining moisture after drying has little or no effect on the fuel. Therefore this FEP is excluded from further evaluation based on the analyses described below. IRSR Issues: CLST1, CLST3, TSPAI4 References: CRWMS M&O 1995b, EPRI 1997, Kohli, R. and Pasupathi, V. 1986, Peehs, M. and Fleisch, J. 1986, NRC 1997, CRWMS M&O 2000b, Knoll and Gilbert 1987. Basis for Screening Decision The fraction of fuel rods with breached cladding is currently estimated to be 0.164% (CRWMS M&O 2000b, Table 13), or about 9.1 rods in an "average" waste package. EPRI 1997 (p. 4-1 and 4-2) gives the BWR and PWR fuel rod reliability as a function of calendar year based and their estimates are lower than that estimated in CRWMS M&O 2000b. Since cask drying operations will remove most of the water from the fuel rods, it is reasonable to expect that the supply of water that remains in the fuel after cask drying will not be sufficient to oxidize the fuel to U3O8. NRC (1997, Section 8.V.1) describes the cask drying criteria with reference to Knoll and Gilbert 1987. Less than 0.43 mol of H2O are expected to be present in a 7-m3 cask after drying. This volume of water produces an insignificant potential for corrosion of the cladding during dry storage or during disposal (CRWMS M&O 1995b, p. 16). Kohli and Pasupathi (1986, p. iii) discusses removal of water from waterlogged fuel rods. Two reactor breached fuel rods were tested, along with two fuel rods that were intentionally defected after irradiation. Since the initial amount of water in the reactor operations breached rods is unknown before drying, a predetermined amount of moisture was added to the intentionally defected rods to enable the extent of the moisture released during the drying to be determined. ANL-WIS-MD-000008 REV 00 22 April 2000 The rods were dried in flowing argon at atmospheric pressure while being heated in a furnace. The center 1.8m of the furnace was heated to 673 K; the remainder was heated to 473 K. The reactor operations breached rods were dried in the as-received condition, then a hole was drilled in the cladding, water was injected, and the experiment was repeated. In the reactor breached rods, the bulk of the uncombined water was removed in 3600 to 4800 sec and all measurable releases ended after 13200 sec (3.7 hours). This set of experiments demonstrated that standard cask drying procedures would remove the water from the failed rods. Peehs and Fleisch (1986, pp. 199-202) described the behavior of waterlogged PWR fuel rods on heating at 400oC in a hot cell. The bulk of the water was released during the cask drying operation. Results of the testing were that the moisture can be removed from defective rods during the cask drying operations and the residual moisture of the cask atmosphere can be minimized. If there were problems with the drying process and rods were filled with water, the extent of the potential corrosion can be estimated. CRWMS M&O (2000b Section 6.3.5) discusses the free volume in a rod and concludes that the average irradiated rod has 17.7 cm 3of free volume although the rod started out with a free volume of 23.3 cm 3. Using the as-manufactured void volume, it is found that there can be no more than 23.3 g of water in one fuel rod. By comparison, 88.2 g of water is required to oxidize the fuel in one fuel rod from UO2 to U 3O8. That is as much water as could be supplied by about 3.8 fully waterlogged fuel rods. In conclusion, waste form degradation from waterlogged fuel rods is excluded from TSPA-SR. There are few failed rods in any WP that could be waterlogged. Experiments have demonstrated that cask drying procedures are effective in removing water. Since the volume of water inside a rod is quite limited, the water could only affect a small amount of fuel. 6.2.2 Cladding Degradation Before YMP Receives It - YMP No. 2.1.02.12.00 FEP Description Certain aspects of cladding degradation occur before the spent fuel arrives at Yucca Mountain. Possible mechanisms include rod cladding degradation during reactor operation, degradation during wet spent fuel pool storage, degradation during dry storage, and rod degradation during shipping (from creep and from vibration and impact) and fuel handling. Screening Decision: Include. Screening Argument: The performance of commercial Zircaloy clad fuel has improved with time. The overall rod failure rate is 0.01 – 0.05 percent. For the TSPA-SR model, CRWMS M&O (2000b, Section 7) gives the CCDF for the fraction of failed rods expected to be received at YMP. In terms of exposing UO2 fuel pellets to the environment after the waste package (WP) fails, most of the fuel that has failed during reactor operation will still have some protection provided by the remaining cladding. After discharge from the reactor, the fuel assemblies are stored in open spent fuel storage pools. An International Atomic Energy Agency (IAEA) survey (IAEA 1988, Table XXVI) reported no evidence of fuel degradation in spent fuel pools and no evidence of significant additional degradation of fuel damaged during reactor operation during residence in the spent fuel pool. The oldest fuel in the survey was Shippingport PWR fuel that has been in wet storage since ANL-WIS-MD-000008 REV 00 23 April 2000 1959. Other fuel reported to have no significant additional degradation during residence in the spent fuel pool has been in wet storage since 1962, 1966, 1968, and the 1970s. The importance of the spent fuel pool storage experience is that fuel failure or significant further degradation is not expected during pool storage, and the fuel failure rates observed from reactor operation are appropriate for the cladding degradation analysis. The DOE has sponsored a Spent Nuclear Fuel Integrity During Dry Storage-Performance Tests Program at the INEEL since 1984. Approximately 26,500 rods have been studied in various commercial dry storage casks. This program also demonstrates what can be expected to happen to fuel in repository WPs during the early thermal period when temperatures are elevated. The cover gas has been monitored and very few rod failures have been observed during dry storage. The importance of the spent fuel dry storage experience (domestic and foreign) is that fuel failure or degradation is expected to be very small during dry storage, and the fuel failure rates are approximated in the cladding degradation analysis by CRWMS M&O (2000b and 2000c). During normal shipping of fuel, no failures have been reported in the literature. Failure probabilities have been calculated for a specific drop test (9-m drop) of a cask with impact limiters, a 0.3-m drop, and normal transport (i.e., vibration). Results show that no yielding of the rods should occur below 63g acceleration. Normal transport would result in accelerations much below these values. IAEA surveyed shipping worldwide and reported in 1988 that there had been no major incidents during 30 years experience connected with irradiated fuel transport. The importance of the fuel transportation and handling experience (domestic and foreign) is in demonstrating that fuel failure or degradation during transportation from vibration or impact is expected to be very small, and that the fuel failure rates are appropriately estimated for the cladding degradation analysis. TSPA Disposition For the TSPA-SR analysis, initial cladding failures from reactor operation, dry storage and transportation has a median value of 0.0948 percent with a range of 0.0155 percent to 1.28 percent for any group (bin) of WPs. In addition, cladding creep failures during dry storage and transportation are integrated into the creep failure analysis for the YMP post-closure clad degradation abstraction. Fuel degradation during spent fuel storage is excluded from TSPA-SR. IRSR Issues: CLST1, CLST3, TSPAI4 References: Einziger, R.E.; Atkin, S.D.; Stellrecht, D.E.; and Pasupathi, V. 1982, Einziger, R.E. and Kohli, R. 1984, EPRI 1997, IAEA 1988, Johnson, A.B., Jr. 1977, Johnson, A.B., Jr.; Bailey, W.J.; Schreiber, R.E.; and Kustas, F.M. 1980, Johnson, A.B., Jr.; Dobbins, J.C.; Zaloudek, F.R.; Gilbert, E.R.; and Levy, I.R. 1987, Manaktala, H.K. 1993, McDonald, S.G. and Kaiser, R.S. 1985, McKinnon, M.A. and Doherty, A.L. 1997, Ravier, G.; Masuy, G.; and Willse, J.T. 1997, Sanders, T.L.; Seager, K.D.; Rashid, Y.R.; Barrett, P.R.; Malinauskas, A.P.; Einziger, R.E.; Jordan, H.; Duffey, T.A.; Sutherland, S.H.; and Reardon, P.C. 1992, Sasaki, S. and Kuwabara, S. 1997, Schneider, K.J. and Mitchell, S.J. 1992, Witte, M.C.; Chun, R.C.; and Schwartz, M.W. 1989, Yang, R.L. 1997, CRWMS M&O 2000b, CRWMS M&O 2000c. ANL-WIS-MD-000008 REV 00 24 April 2000 Basis for Screening Decision 6.2.2.1 Pin Degradation During Reactor Operation - YMP No. 2.1.02.12.01 The performance of commercial Zircaloy-clad fuel has improved with time. Early cores (up to 1985) had a rod failure rate averaging 0.02 percent to 0.07 percent (EPRI 1997, p. 4-1). After 1985, the rod failure rate decreased to 0.006 percent to 0.05 percent. The overall rod failure rate for both time periods is 0.01 percent to 0.05 percent. Manaktala (1993, p. 3-4) shows fuel reliability as a function of calendar year for both PWR and BWR fuel from 1969 through 1976. He shows early PWR cores had failure rates over 1 percent but the rates dropped below 0.1 percent by 1973. The BWR failure rates fell below 0.1 percent after 1975. Yang (1997, Table 2, p. 10), summarized the frequency and type of assembly failures from 1989 through 1995. Yang’s failure data are for assemblies, and can be converted to rod failure rates using 2.2 failed rods per failed assembly (EPRI 1997, p. 4-1), and an average of 221 rods per assembly. Occasionally a specific core will have a higher failure rate. These data are included in the EPRI fuel failure rate, and affect the averages slightly. After a steam generator replacement in 1981, one reactor operator reported 0.26 percent of the rods (32 percent of the assemblies) were damaged by fretting from foreign particles, such as debris from the steam generator replacement (McDonald and Kaiser 1985, pp. 2-4, 2-5). This case of severe core damage shows that a relatively small fraction of the rods was actually damaged. The French ( Ravier et al. 1997, Figure 4, p. 34) report annual rod failure rates from 5.5´10-4 to 0.0 from 1986 through 1996. Their reactors are PWRs based on the Westinghouse design, and have similar fuel designs as the United States reactors. The Japanese (Sasaki and Kuwabara 1997, pp. 13-14) report failure rates of 0.01 percent for BWRs and 0.002 percent for PWRs. Their BWR design is based on the General Electric (GE) design, and their fuel designs are similar to the United States reactors. Both rates are similar to the current United States observed rates. Rod damage characteristics in reactor operation (EPRI 1997, pp. 4-2, 4-3) are: Pinhole and through wall hairline cracks 80 to 90 percent Intermediate condition 10 to 20 percent Severe damage 0.04 to 0.9 percent In terms of exposing UO2 fuel pellets to the environment after the WP fails, most of the fuel that has failed during reactor operation will still have some protection provided by the remaining cladding. Reactor operation failed rods are included in TSPA-SR. The details of the analysis are presented in CRWMS M&O 2000b, which also develops the stress distribution used in the creep analysis. 6.2.2.2 Pin Degradation During Spent Fuel Pool Storage -YMP No. 2.1.02.12.02 Rod failure during spent fuel pool storage is excluded from TSPA-SR. The importance of the spent fuel pool storage experience is that fuel failure or further degradation of rods already failed during reactor operation is not expected during pool storage, and the fuel failure rates observed from reactor operation, dry storage, and shipping are appropriate for the cladding degradation analysis. ANL-WIS-MD-000008 REV 00 25 April 2000 After discharge from the reactor, the fuel assemblies are typically stored in spent fuel storage pools. An International Atomic Energy Agency (IAEA) survey (IAEA 1988, Table XXVI) reported no evidence of fuel degradation in spent fuel pools and no evidence of significant additional degradation of fuel damaged during reactor operation. The oldest fuel in the survey was Shippingport PWR fuel that has been in wet storage since 1959. Other fuel reported to have no further degradation has been in wet storage since 1962, 1966, 1968, and the 1970s. Under the DOE Spent Fuel and Fuel Pool Component Inventory Program, the effect of storing both fuel with intact cladding and fuel with failed cladding has been studied. An international survey of in-water storage (Johnson et al. 1980, p. iii) reports no cases of fuel cladding degradation during pool storage. Johnson (1977, p. 20), reports: “Operators at several reactors have discharged, stored, and/or shipped relatively large numbers of Zircaloy-clad fuel which developed defects during reactor exposures, e.g., Ginna, Oyster Creek, Nine Mile Point, and Dresden units I and II. Several hundred Zircaloy-clad assemblies which developed one or more defects in-reactor are stored in the GE-Morris pool without need for isolation in special cases. Detailed analysis of the radioactivity in the pool water indicates that the defects are not continuing to release significant quantities of radioactivity.” 6.2.2.3 Pin Degradation During Dry Storage -YMP No. 2.1.02.12.03 Creep strain damage and pin damage (failure) during dry storage are included in the creep degradation analysis described in CRWMS M&O (2000c, Section 6.2). It is assumed that all fuel undergoes 20 years of dry storage starting at 350oC. The amount of creep strain during dry storage is calculated and added to the creep from shipping and the YMP thermal transient (CRWMS M&O 2000c, Figure 5). In addition, 0.045 percent of the rods are assumed to have failed in dry storage from handling and other causes (CRWMS M&O 2000b, Table 13). The DOE has sponsored a Spent Nuclear Fuel Integrity During Dry Storage-Performance Tests Program at the INEEL since 1984 (McKinnon and Doherty 1997, pp. 2.1, 5.16). Approximately 26,500 rods have been studied in various commercial dry storage casks. This program also demonstrates what can be expected to happen to fuel in repository WPs during early storage times when temperatures are elevated. The dry storage cask fuel cover gas has been monitored and very few fuel rod failures have been observed during dry storage. During fuel consolidation, approximately 10 rods are believed to have developed small leaks. Some rods leaked, but the release was over a period of approximately two months, a very slow gas release. Only 0.5 percent of an individual rod’s 85Kr content was released from the leaking rods. After shipment to the INEEL, the observed rod failure was similar to that after reactor operation. For rods that were not consolidated, 2 rods leaked out of 16,700 rods producing a failure probability of 1.2 ´ 10-4 per rod, below the 0.045 percent failure rate used in TSPA-SR for failure in dry storage. Some utilities experimented with fuel consolidation before the introduction of dry storage facilities. After consolidation at INEEL (McKinnon and Doherty 1997, pp. 2.1, 5.16), 12 rods failed out of 9800 (failure probability =1.22 ´ 10-3), an increase by 1 order of magnitude from the unconsolidated failure rate. Although no utilities are currently using consolidation, the rod failure rate on 0.045% used in TSPA-SR for dry storage contains a component of the failure rate from consolidation. Dry storage tests were performed at the Nevada Test Site (Johnson et al. 1987, p. iv) with 17 spent fuel assemblies, each in an individual test. These tests contained 3468 rods, and cladding temperatures varied from 168 to 27X°C. One of the fuel rods failed during these tests. This assembly was exposed to air at 275'C and had nine thermal cycles. The estimated hole size was 1 pm. No further degradation was observed in this one failed rod after the initial failure. No visible damage was observed in the other tests. Accelerated high temperature tests were performed on 15 rods (Einziger et al. 1982, pp. 65, 69). Post irradiation studies of failure mechanisms of well-characterized pressurized water reactor rods were conducted for up to a year at 482'C, 510°C, and 571°C in limited air and inert gas atmospheres. No cladding breaches occurred. The cladding had crept away from the pellets and showed a smoother profile. Strains from 1.7 percent to 7 percent were measured (their Table IV). One rod had a local creep strain as high as 12 percent (their Figure 3, p.69). The extended lifetime is attributed to significant creep strain of the Zircaloy cladding, which decreases the internal rod pressure. The cladding creep also contributes to radial cracks through the external oxide layer and internal fuel-cladding chemical interaction layers (layer of a few micrometers where some zirconium/U02 interaction occurs), which propagated into and arrested in an oxygen stabilized alpha-Zircaloy layer. Since cracks extended only for a few micrometers and were arrested, their significance was small. There were no signs of either additional cladding hydriding, stress corrosion cracking (SCC), or fuel pellet degradation. A second series of tests (Einziger and Kohli 1984, pp. 107 and 114, Table 111) was performed on five PWR spent fuel rods. They were pressurized to a hoop stress in the range of approximately 145 to 155 MPa, for times up to 2101 hours at 323'C. The conditions were chosen for limited annealing of in-reactor irradiation hardening. With the stresses in the range of 145 MPa, creep of 0.004% to 0.16% was observed. No cladding breaches occurred, although significant hydride agglomeration and reorientation took place in one rod that cooled under stress. Einziger and Kohli (1984, pp. 107 and 114, Table 111) state that these high-temperature tests based on creep rupture as the limiting mechanism indicate that storage at temperatures between 400 and 440°C may be feasible for annealed rods. Schneider and Mitchell (1992, p. 2.7) summarized experience in the foreign dry storage programs. At that time, seven countries had some fuel in dry storage. They conclude that Light Water Reactor (LWR) fuel can be stored for up to 100 years at temperatures of 320 to 400°C in an inert atmosphere and, if exposed to air, will last comparable times if the temperature is limited to 135 to 160°C. The Canadians have large quantities of irradiated fuel with Zircaloy cladding that has been exposed to air with favorable results. For over eight years, they tested the effects of exposing fuel with defective cladding to moist and dry air with favorable results (no observed strain from U02 oxidation nor cladding failure propagation). The importance of the spent fuel dry storage experience (domestic and foreign) is that fuel failure or degradation is expected to be very small during dry storage for the current fuels. It is also important to note that under repository conditions, most of the cladding creep or DHC failures would be expected to occur in the first 100 years after closure, this is the time period ANI-WIS-MD-000008 REV 00 26 Apr112000 when fuel temperatures are highest and conditions are closest to dry storage conditions. Dry storage is considered to be a good representation of this period. 6.2.2.4 Pin Degradation During Fuel Shipment and Handling - YMP No. 2.1.02.12.04 During normal shipping of fuel, no failures have been reported in the literature. Sanders et al. (1992, Table 111-10, p. 111-137), analyzed transportation accidents. Failure probabilities were calculated for a specific drop test of nine meters (9-m), a 0.3-m drop, and normal transport (i.e., normal vibration). The 9-m drop was by far the most severe, with failure probabilities on the order of 2 x The normal transport failure is 2 x per rod. This reference generated acceleration versus frequency curves for trucking and rail shipping. It then looked at structural damage from a sampling of the hazard curves and the structural analysis and concluded that no additional damage is done in shipping. This value is conservatively bounded by a failure fraction of 1 x 10" to account for other transport accident conditions and is included in the TSPA-SR. In addition, it was assumed that all rods undergo three (3) weeks of transportation at a temperature of 35OoC when considering the creep strain damage. This inclusion is described in CRWMS M&O (2000b, Section 6.10.1) and (2000c, Section 6.2.1). Witte et al. (1989) performed an analytical evaluation of the potential impacts of all transport movements. They concluded (Witte et al. 1989, p. 194, Table 3) that no yielding of the rods would occur below 63g acceleration. Normal transport would result in acceleration far below these values. IAEA (1988, p. 114) surveyed shipping worldwide and reported that "to date, there have been no major incidents during 30 years experience connected with irradiated fuel transport." The importance of the fuel transportation handling experience (domestic and foreign) is in demonstrating that very little fuel failure or degradation is expected during transportation from vibration or impact, and that the fuel failure rates observed from reactor operation are appropriate for the cladding degradation analysis. 6.2.3 General Corrosion of Cladding - YMP No. 2.1.02.13.00 FEP Description General corrosion of cladding could expose large areas of fuel and produce hydrides. Screening Decision: Exclude, low consequence Screening Argument: General corrosion is synonymous with zirconium oxidation. The outer surface of the cladding becomes oxidized with ZrOa providing a protective film which adheres to the surface and slows down further oxidation. The oxidation could be from 0 2 consumption (dry oxidation) or Hz0 consumption (wet oxidation) and the dry and wet oxidation rates are very similar. For the fuel in the repository, oxidation does not occur until the WP is penetrated. Various analyses have concluded that cladding oxidation under repository conditions would not lead to rod failure. The amount of oxide film resulting from general corrosion as a function of time since WP failure has been calculated but is insignificant if the WP remains sealed for the first 100 years. There was no measurable degradation of the Zircaloy cladding after 21 years of in-water spent fuel pool storage times. ANI-WIS-MD-000008 REV 00 27 Apr112000 TSPA Disposition Exclude, low consequence IRSR Issues: CLST1, CLST3, TSPAI3, TSPAI4 References: Bradley, E.R.; Bailey, W.J.; Johnson, A.B., Jr.; and Lowry, L.M. 1981, Einziger, R.E. 1994, Hillner, E.; Franklin, D.G.; and Smee, J.D. 1998, IAEA 1998, Rothman, A. J. 1984 Basis for Screening Decision General corrosion is synonymous with zirconium oxidation. The outer surface of the cladding becomes oxidized with a Zr02 film, which adheres to the surface and slows down further oxidation. The oxidation could be from 0 2 consumption (dry oxidation) or Hz0 consumption (wet oxidation). For the fuel in the repository, this does not occur until the WP is penetrated. The effect of surface oxidation is threefold. The oxidation could thin the cladding, contributing to cladding failure by creep rupture. Wet oxidation generates hydrogen, and some of the hydrogen is absorbed into the cladding to form hydrides. This could lead to delayed hydride cracking (DHC), or general hydride embrittlement. In the extreme, the oxidation could lead to cladding disintegration and expose the fuel pellets to the WP environment. Wet oxidation has been studied for over forty years because of its importance in reactor operation. Rothman (1984, pp. 6 - 13) discusses cladding oxidation in repository conditions in great detail and his Table 3 compares the predicted cladding loss using six different oxidation correlations and predicts cladding thinning of 4 to 53 pm after 10,000 years at 180°C (a conservative temperature assumption) which would not lead to rod failure. Recently, Hillner et al. (1998, p. 9) published a recommended Zircaloy corrosion correlation based on Bettis Atomic Power Laboratory experiments. Bettis developed Zircaloy for Navy reactors in the early 1950s and has an extensive data base on Zircaloy performance. They have continuous autoclave corrosion tests on some samples for 30 years. Some samples have developed oxide thickness as great as 110 pm, greater than those expected during repository corrosion. The experiments show that the corrosion rate is determined by diffusion of oxygen ions through the corrosion film. This corrosion film is generated in layers, with the lower layer staying very consistent. The consistency of the lower 2 pm of oxide film leads to a steady corrosion rate after a transition period. The recommended post-transition rate equation is: ATh =1.69 x l o 9 x exp(-111452lT) (Eq. 1) where: ATh = oxide growth rate, pm/y T = temperature, K This correlation rate equation is similar to the equations developed by others, but predicts a slightly higher corrosion rate. It is taken from Equation 7 of Hillner at al. (1998), doubled for a conservative correction for irradiation (Hillner et al. 1998, pp. 6, 9) and converted to ANI-WIS-MD-000008 REV 00 28 Apr112000 micrometers per year (lpm = 14.7 mg/dn?) (IAEA 1998, p. 178). The pre-transition rate is slower than the post-transition rate. The effect of irradiation conditioning before beginning the corrosion is to accelerate the corrosion rate for a few micrometers. To be conservative, Hillner et al. doubled the corrosion rate for all time. The correlation shows a strong Arrhenius temperature relationship with the corrosion rate becoming very small below 200°C. This is consistent with the data from Hillner et al. Corrosion tests at 270°C for 8.2 years have produced approximately 4 pm of oxide, while corrosion tests at 360°C have produced films 88 pm thick in 7.8 years (Hillner et al. 1998, p. 25). For general cladding corrosion, the WP must be breached. The amount of oxide film from general cladding corrosion as a function of time since WP failure has been calculated. These calculations were carried out from the WP failure time to one million years. The amount of cladding that is consumed is approximately 57 percent of the oxide thickness because of the volume increase associated with Zircaloy oxidation. The cladding oxidation calculation was done for the design bases (hot) rod in the hottest repository region. If the WP is not breached after 100 years from emplacement, little general corrosion is expected. Hillner et al. (1998, Figure 5) compare the weight gain of the samples in water which correlates to the corrosion rate in water (the correlation used here) with that of steam. The steam corrosion rate is about 30 to 40 percent slower. A steam environment is expected to last for 1000 to 5000 years in the near field. Einziger (1994, p. 556, Equation 14) states that dry oxidation of zirconium is slightly slower than the wet corrosion rate. The conclusion from CRWMS M&O (1998, Section 6.3.1.1.4.1), is that general cladding corrosion is not a problem for fuel cladding in the average WP but could be a problem for fuel cladding in the design WP if the design WP fails before 100 years. Bradley et al. (1981, p. 38) performed metallurgical examinations of Zircaloy-clad fuel rods from two bundles (0551 and 0074) of the Shippingport PWR Core 1 "blanket" fuel after extended inwater spent fuel pool storage (21 years for 0551, and 16 years for 0074). The oxide film thickness on the Shippingport fuel rods after reactor operation was reported to be an average cladding oxide film thickness of 1.8 pm (0551) and 2.4 pm (0074). After extended in-water spent fuel pool storage, the average cladding oxide film thickness was found to be 1.7 pm (0551) and 2.3 pm (0074)(Bradley et al. 1981, p. 38). The slight disagreement in these values is attributed to differences in measurement technique and experimental error. These results led to the conclusion that no significant change in oxide thickness occurred even after 16 to 21 years of pool storage. This conclusion is further supported by the observation that Zircaloy tube sheets (that had been cut to remove bundle 0551 fuel rods in 1960) stored in water for over 20 years were unblemished and showed no evidence of reaction with water. In summary, the conclusion in this AMR is that general cladding corrosion is not a problem for fuel cladding in the average WP but could be a problem for fuel cladding in the design WP if the WP fails before 100 years. 6.2.4 Microbiologically Influenced Corrosion (MIC) of Cladding - YMP No. 2.1.02.14.00 FEP Description Microbiologically Influenced Corrosion (MIC) of cladding potentially may be a local cladding corrosion mechanism where microbes produce a local acidic environment that could produce multiple penetrations through the fuel cladding ANI-WIS-MD-000008 REV 00 29 Apr112000 Screening Decision: Exclude, low probability. Screening Argument: There is no indication that MIC occurs on zirconium metal or alloys. The two major forms of MIC for materials being considered for waste packages are (1) sulfide attack through the action of sulfate reducing bacteria (SRB) and (2) corrosion induced by organic acids secreted from certain bacteria. With respect to these forms of MIC: (1) SRB do not affect zirconium, and (2) corrosion induced by organic acids is unlikely because of zirconium's tolerance to organic acids and a wide range of pHs. TSPA Disposition Exclude. IRSR Issues: CLST1, CLST3, TSPAI3, TSPAI4 References: Little, B. and Wagner P. 1996, Yau, T.L. and Webster, R.T. 1987, Hillner, E.; Franklin, D.G.; and Smee, J.D. 1998, McNeil, M., and Odom, A. 1994, CRWMS M&O 2000a Basis for Screening Decision The term microbiologically influenced corrosion (MIC) is commonly used to designate corrosion caused by the presence and activities of microorganisms at the surfaces of metals. Little and Wagner (1996, p. 367) published an overview of microbiologically influenced corrosion of metals and alloys used in the storage of nuclear wastes. They indicate that MIC is a form of localized corrosion that results in pitting, selective leaching, crevice corrosion, underdeposit corrosion, and enhanced erosion/corrosion. Little and Wagner (1996, pp. 367-368) describe several mechanisms for microbiologically influenced corrosion. In addition, various case studies are presented that document microbiologically influenced corrosion of alloys of iron, nickel, and copper. However, it should be noted that there is no indication in the literature that MIC occurs on zirconium metal or alloys. Yau and Webster (1987, p. 709) report that no corrosion of zirconium metal from marine organisms was found during sea water corrosion tests for 129 days. Hillner et al. (1998, p. 11) indicate that there are two major forms of MIC for materials being considered for WPs. They are (1) sulfide attack through the action of sulfate reducing bacteria and (2) corrosion induced by organic acids secreted from certain bacteria. With respect to attack by SRB, Hillner et al. 1998 reference the work of McNeil and Odom (1994, p. 176). McNeil and Odom (1994, p. 176) indicate by thermodynamic calculations that SRB do not affect zirconium. With respect to corrosion induced by organic acids, Hillner et al. (1998, p. 11) noted that it is most unlikely because of zirconium's tolerance of a wide range of pHs and that it is unlikely that production of weak organic acids will have an adverse effect on the passivation of Zircaloy by a Zr02 film. Yau and Webster (1987, p. 717) also note that zirconium resists a wide range of organic compounds, including acetic acid, acetic anhydride, formic acid, urea, ethylene dichloride, formaldehyde, citric acid, lactic acid, oxalic acid, tannic acid, and trichloroethylene. CRWMS M&O 2000a summarizes the corrosion potential of zirconium to many chemicals. ANI-WIS-MD-000008 REV 00 6.2.5 Acid Corrosion of Cladding From Radiolysis - YMP No. 2.1.02.15.00 FEP Description Radiolysis in a nitrogenloxygen gas mixture with the presence of water film results in the formation of nitric acid (HNO3) and hydrogen peroxide (H202) ions that can enhance corrosion of the fuel cladding. Screening Decision: Exclude, low probability. Screening Argument: Radiolysis has the potential to produce chemical species that may not otherwise be anticipated in the repository. One example is the production of nitric acid as a result of the radiolytic reaction between nitrogen and oxygen with the production of nitrogen oxides and subsequent formation of nitric acid. Another example is the production of hydrogen peroxide. An analysis of the gamma and neutron dose at 1000 years at the WP surface, thus well before the first anticipated breach of the waste packages, provides a total dose of 0.466 remlhour at the waste package surface (CRWMS M&O 1999d, p. 17). The dose may be too low for the production of sufficient peroxide or nitric acid to cause a corrosion concern. Zirconium is used for fuel cladding because of its high resistance to corrosion in high radiation fields. Yau and Webster 1987 (p. 710) note that zirconium has excellent corrosion resistance to nitric acids. They also report that zirconium does not corrode in concentrated hydrogen peroxide. TSPA Disposition Exclude. IRSR Issues: CLST1, CLST3, TSPAI3, TSPAI4 References: CRWMS M&O 2000a, CRWMS M&O 1999d, Hansson, C.M. 1984, Van Konynenburg, R.A.; Curtis, P.G.; and Summers, T.S.E. 1998, Yau, T.L. and Webster, R.T. 1987, IAEA 1993, Hillner et al. 1998. Basis for Screening Decision It is recognized that radiolysis may create various chemical species that may not otherwise be expected in the repository. Examples are the production of nitric acid as a result of the radiolytic reaction between nitrogen and oxygen and the formation of hydrogen peroxide. In either case, accelerated corrosion of Zircaloy would not be predicted in the absence of irradiation since zirconium alloys have been shown to be relatively inert in both media as discussed in Yau and Webster 1987 and CRWMS M&O (2000a, Sections 6.1.6 and 11.4). As one example, zirconium equipment is used in the chemical processing industry where peroxide strengths of 90% are used. The service life has been increased by an order of magnitude compared to graphite components previously used, which were generally considered to be very inert. In nitric acid, zirconium and its alloys are inert up to acid concentrations of 65%. Provided radiolysis does not produce nitric acid at greater concentrations, there will be no impact on the uniform corrosion rate as a result of nitric acid production from radiolysis. Radiolysis was reviewed in detail in the IAEA 1993 report (pp. 82-92). Specifically the topic of radiolysis of "thick oxide film effects" was discussed. The report noted (IAEA 1993, p. 91) that "the radiation enhancement of Zircaloy corrosion in the aqueous phase arises from the ANI-WIS-MD-000008 REV 00 31 Apr112000 synergistic interaction of radiation and water chemistry. It has been recognized that in the early stages of Zircaloy corrosion the acceleration by reactor radiation, which is usually observed in oxygenated water, is suppressed in the presence of excess hydrogen in the aqueous phase, but that beyond the threshold oxide thickness Zircaloy specimens exposed to low-oxygen water corrode at accelerated rates under irradiation as if they were immersed in oxygenated water." This implies that under the worst conditions of high radiation levels and excess oxygen ions, the corrosion rate is only accelerated by a factor of three. However, an analysis of the gamma and neutron dose in the repository at 1000 years, thus well before the first anticipated breach of the waste package, predicts a total dose of 0.466 remhour at the waste package surface (CRWMS M&O 1999d, p. 17). This dose is significantly lower than that in an operating reactor and may in fact be too low to produce sufficient peroxide to produce an accelerated corrosion. In the unlikely event that premature failure of a WP should occur, the waste package temperature would preclude the presence of liquid water, and thus the presence of peroxide, for at least 1000 years. At this time the surface radiation dose will be very low level as noted above, so that the practical likelihood of producing peroxide is extremely low under any conditions. Further support for this position can be obtained from Hillner et al. 1998. On page 12, it is stated that "The corrosion rate of Zircaloy in water is limited by the slowest of the basic corrosion steps: dissociation of water into oxygen and hydrogen ions, diffusion of oxygen ions through the oxide film, oxidation of the Zr metal, diffusion of electrons through the oxide, and hydrogen-ion reduction by the electrons at the water-to-oxide interface. At temperatures for current testing, that is, above about 45OoF, diffusion of oxygen ions through the oxide film is rate limiting. In extrapolating to lower temperatures, one of the other basic steps could become rate limiting, which would decrease the corrosion rate to less than predicted by the extrapolation". In summary, if corrosion is controlled by the oxygen ion diffusion rate, the presence of peroxide would have no effect on Zircaloy corrosion and the corrosion rate would be expected to be that of uniform corrosion. Yau and Webster (1987, pp. 707-721) review the corrosion of zirconium under various chemical environments for commercial applications (corrosion rates and time scales of interest for industrial applications, not for repository time scales). Zirconium is resistant to corrosion from HC1 to temperatures well above boiling (Yau and Webster 1987, p. 710). In basic solutions, Hansson (1984, summary page) measured corrosion of Zircaloy 2 in anaerobic cement pore solutions of DHS of 12.0 to 13.8. She concludes: "Thus. it mav be concluded that active corrosion of Zircaloy 2 in anaerobic concrete will not occur and by comparison with measurements on steel, it is likely that the passive corrosion rates will be even lower in concrete than those measured in the synthetic pore solution." Yau and Webster 1997 report no corrosion in seawater, brackish water, and polluted water. Zirconium is resistant to corrosion from sulfuric acids in concentrations less than 20 percent and corrodes slowly in terms of commercial applications (not repository time scales) in sulfuric acid solutions below 65 percent. Van Konynenburg et al. (1998, p. 7-17) performed container material scoping tests using Zircadyne 702 (a Zirconium hafnium alloy) in 0.01 mol/L each of sodium formate (NaCOOH), nitric acid (HNO?), NaCI, H202, and 0.02 mol/L sodium oxalate (NazCzOd), with an initial pH of approximately 4, temperature of 9Of 5 "C, and typical test duration of 96 hours. The test solution was designed to represent highly concentrated forms of acids formed by radiolysis. Formic acid, ANL-WIS-MD-000008 REV 00 32 April 2000 a potential pH buffer, was not present in the Zircadyne 702 test. They report a final pH of 4.26 and corrosion rate of 50 pmlyr for zirconium in the acid solution. The duration of highly alkaline conditions in an emplacement drift depends on the assumed use of concrete. If concrete (emplacement drift liner material) is not used, the duration of highly alkaline conditions in the emplacement drift is less than 5000 years, with the long-term pH at about 8. If concrete material is used in the liner, the expectation is that the pH in the emplacement drift will be quite basic for the first 10,000 years. After emplacement of the waste packages, the radioactive decay of the waste will heat the drifts and disturb the normal percolation of water through the mountain. As the drifts cool, some of the ground water percolating through the mountain may drip into the drifts and some of the waste packages. Through time, the components of the engineered barrier are expected to degrade. The metallic materials of the WPs, pedestals, and drip shield are expected to undergo humid air and aqueous corrosion, and the concrete materials, if used, are expected to undergo hydrothermal degradation. As the materials deteriorate, the drift liner will fall onto and around the WP and the pedestals will collapse lowering the WPs toward the floor. The WPs and drip shield are expected to undergo humid corrosion, and if wetted by dripping or high relative humidity, to undergo aqueous corrosion. It is assumed that corrosion degradation of Zircaloy fuel cladding does not begin until the WP is breached (typically 10,000 years or more after emplacement). A steam environment is expected to last for 1000 to 5000 years in the near field. Once a WP is breached, water may enter the WP as water vapor or as drips. The WP and fuel will have cooled down by that time to an extent that will allow a water film to form on the WP surface and on the fuel cladding that is thick enough to support corrosion reactions. The radiation dose would also be a lowered by that time to a level that is considered insignificant to long-term engineered barrier performance. In summary, the WPs are expected to be intact for thousands of years after emplacement. During the first 1000 to 5000 years, the environment around the WPs is expected to be steam. This environment will prevent nitric acid or hydrogen peroxide from pooling on or coating the WP (or fuel cladding) and would thus control acid corrosion. 6.2.6 Localized Corrosion (Pitting) of Cladding - YMP No. 2.1.02.16.00 FEP Description Localized corrosion in pits could produce penetrations of cladding. Screening Decision: Exclude, low probability. Screening Argument: Corrosion tests are currently being performed for repository type chemicals, but results are not yet available. The probability of pitting, crevice corrosion, or some other form of localized corrosion depends on the local water chemistry and for predicted WP water chemistry is unlikely. Local corrosion of zirconium-based clad by pitting is observed to occur when: 1. Chloride levels exceed 1 mmol/L; 2. Inhibitor ions, such as nitrate, sulfate, and bicarbonate are absent; 3. Ferric ion levels are high and the pH low; ANI-WIS-MD-000008 REV 00 33 4. The passivating ZrOz film is thin, absent, or mechanically disrupted. Note that the passivating film is somewhat soluble at moderately low pH (pH < 3.5), though its dissolution, and that of the underlying metal, is primarily observed at very low pH. Since none of these conditions is predicted to occur inside the waste package, the process is unlikely. TSPA Disposition Exclude ZRSR Issues: CLST1, CLST3, TSPAI3, TSPAI4 References: Yau, T.L. and Webster, R.T. 1987, Hansson, C.M. 1984, Rothman, A.J. 1984, Hillner, E.; Franklin, U.G.; and Smee, J.U. 1998, Yau, 1'. L., 1984, Van Konynenburg, R.A.; Curtis, P.G.; and Summers, T.S.E. 1998, Cragnolino, G. and Galvele, J.R. 1978, Maguire, M. 1984, Krauskopf, D.K. and Bird, D.K. 1995, Uziemblo, N.H. and Smith, H.D. 1989, Greene et al. 2000, CRWMS M&O 2000f. Basis for Screening Decision Chloride-accelerated corrosion is neglected, in part, because the water in the vicinity of the repository (represented by water from well 5-13 and referred to as J13 water) contains only 0.2 mmol1L chloride. This is roughly 115 of the amount required to cause pitting in solutions otherwise free of corrosion inhibitors, though evaporative concentration might in some cases raise 513 chloride levels. Yau and Webster (1987, pp. 707-721) review the corrosion of zirconium under various chemical environments for commercial applications (their corrosion rates and time scales are of interest for industrial applications, not for repository time scales). CRWMS M&O 2000a also review the potential for corrosion of Zircaloy cladding. Zirconium is resistant to corrosion from HC1 to temperatures well above boiling (Yau and Webster 1987, p. 710), basic solutions, sea water, brackish water, and polluted water. It is resistant to corrosion from sulfuric acids in concentrations less than 20 percent and corrodes slowly in terms of commercial applications (not repository time scales) in solutions below 65 percent. Zirconium is susceptible to corrosion from fluoride ions (> 100 ppm and pH dependent), hydrofluoric acid, and aqua regia (mixture of hydrochloric and nitric acid). Yau and Webster (1987, p. 717) report that zirconium is among the structural metals that are most resistant to crevice corrosion. If the zirconium oxide layer is intact, it is not susceptible to galvanic corrosion if coupled to a more noble metal (platinum, gold, graphite, titanium, and silver). It has a corrosion rate of 18 pdyr for unwelded specimens in 3.5 percent boiling NaCl at a pH of 5 but no observed corrosion at a pH of 6. Hansson (1984, summary page) reports no corrosion rate increase in anaerobic cement pore solutions of pHs of 12.0 to 13.8. It is susceptible to pitting from FeC13 ions (CRWMS M&O 2000a, Section 6.2.2.1). FeC13 pitting requires femc ions, which are the dominant species at pH less than 3. Current analysis of the water chemistry inside the WP predicts pHs greater than 3.6 (CRWMS M&O 2000f, Table 3). Rothman (1984, pp. 4-5) qualitatively lists the chemical components of J-13 water and the ability of these chemicals to attack zirconium. He concludes that Zircaloy corrosion from these chemicals is unlikely. Hillner et al. (1998, p. 11) also qualitatively compared Zircaloy corrosion ANL-WIS-MD-000008 REV 00 34 April 2000 potential with J-13 components and concluded that corrosion the J-13 water, in the measured concentrations and pH levels, would not have a detrimental effect on the zirconium corrosion behavior. The possibility of chloride-accelerated corrosion is made more unlikely by the fact that J-13 water contains anions that are known to inhibit pitting. Nitrate at 0.16 mmolIL, sulfate at 0.25 mmollL, and bicarbonate at 2 mmollL are known inhibitors at these levels unless chloride levels are raised to - 1 mol1L (Greene et al. 2000). Ferric and cupric ions can cause pitting of Zircaloy. Although large amounts of ferric iron are expected to accumulate through degradation of breached waste packages, very little of it will remain in solution due to the low solubility of Fe(II1)-containing solids. Ferric ion levels are not predicted to exceed 4 x mmol1L. The predicted Fe levels are many orders of magnitude less than those that are observed to accelerate pitting (Greene et al. 2000). Moreover, ferric ion pitting typically requires pH below 2.5, well below the lowest value predicted to prevail in the WP. Passivating surface layers on spent nuclear fuel rods tend to be thick and well-developed; 5 to 100 pm for PWRs and 5 to 20 pm for BWRs, malung them more resistant to pitting than solids used in many experiments (CRWMS M&O 2000a). Published results (Yau 1984, pp. 14011-14018) show general, crevice and stress corrosion cracking (SCC) tests with zirconium in concentrated acids to simulate concentrated geothermal salts. In terms of chemical composition, these tests are expected to be more severe than expected at YMP, although the test time periods are very short. Yau (1984) also looked at crevices and SCC with mixed salts. The solution contained 25% NaCl + 0.5% CH3COOH + 0.1% S t saturated H2S, with or without saturated C02. Both welded and non-welded samples were exposed in U-bend SCC tests, and general and crevice corrosion tests. Tests were conducted both at room temperature and under high temperaturelpressure autoclave conditions. The pH was between 2.3 and 3 (measured at room temperature), and the test duration was for 30 or 60 days. The U-bend samples showed no cracking after 60 days. Steel coupled U-bend samples showed no hydrogen pickup, although the steel nuts and bolts corroded badly during the testing. These test solutions became contaminated with iron, but the iron contamination resulted in no degradation of zirconium's corrosion resistance. Crevice tests showed no crevice corrosion. Autoclave tests of general corrosion showed negligible corrosion, with "slight" weight gain of the samples (the rate was not given in the paper). The author also noted that fluoride pitting is not expected in geothermal fluids, because the presence of silica tends to trap the fluoride and form non-corrosive complex ions. Van Konynenburg et al. (1998, p. 7-17) performed certain material scoping tests using Zircadyne 702 (a Zirconium hafnium alloy) in 0.01 mol1L each of sodium formate (NaCOOH), nitric acid (HNO3), NaC1, H202, and 0.02 mol1L sodium oxalate (NazCzOd), with an initial pH of approximately 4, temperature of 9Of5 "C, and typical test duration of 96 hours. The test solution was designed to represent highly concentrated forms of acids formed by radiolysis. Formic acid, a potential pH buffer, was not present in the Zircadyne 702 test. They report a final pH of 4.26 and corrosion rate of 50 pmlyr for zirconium in the acid solution. ANI-WIS-MD-000008 REV 00 35 The data presented in the papers by Cragnolino and Galvele (1978, p. 1056) and Maguire (1984, p. 177) show that Zircaloy could pit in solutions with chloride ions (1 mollL or higher and at room temperature). Cragnolino and Galvele studied zirconium and Zircaloy-4, but Maguire only studied zirconium and data on zirconium may not be directly applicable to Zircaloy-2 and -4. These experiments were performed in air but without any additional oxidants present. Maguire reports that the critical pitting potential values are not affected by pH from 0 to 12.8 or by temperature. He also argues that presence of oxidizing impurities such as Fe3+ and cu2+ would make Zirconium more susceptible to localized corrosion. Maguire (1984, p. 188) reports: "Our experience has been that if the pH is maintained above three no pitting occurs." A pH less than 3 is required for the preferential formation of the Fe3+ ions (Krauskopf and Bird 1995, p.228, Figure 9-2) and the pH inside the WP is predicted to stay above 3.6 (CRWMS M&O 2000a. Table 3). Zircaloy is susceptible to pitting with F ions. Uziemblo and Smith (1989, pp. 10-19) measured corrosion rates at various pH and F ion concentrations. With 100 ppm F ions, they found that Zircaloy-4 corroded a negligible amount at pH > 8, 32 pmlyr at pH = 6 and 1 mmlyr at pH = 5. Yau (1984, p. 140/3), states that silica tends to trap the fluoride and to form non-corrosive complex ions. The experiments by Uziemblo and Smith (1989) show the importance of pH on the corrosion rate. It also shows the importance of estimating the composition of the water and the degree of concentration for the water that contacts the fuel, since the fluoride content of 5-13 is about 2.4 ppm. Experiments with 500 ppm cu2+ in 3.5 percent boiling NaCl also show a strong pH dependency with a corrosion rate of 18 pmlyr at pH of 5 and nil at a pH of 6 and above (Yau and Webster 1987, pp. 716-717, Table 13). CRWMS M&O 2000f calculated the chemistry in the WP after WP failure. The analysis predicted that the water would be basic for most of the time but could turn acidic for a short period because of the sulfur within the carbon steel. The pH for all the cases studied was always above a pH of 3.6 (CRWMS M&O 2000f, Tables 2 to 5). Therefore water chemistries that would permit pitting are not expected. 6.2.7 Localized Corrosion (Crevice Corrosion) of Cladding - YMP No. 2.1.02.17.00 FEP Description Localized corrosion in crevices could produce penetrations of cladding. Screening Decision: Exclude. Screening Argument: Zirconium is very resistant to crevice corrosion. CRWMS M&O 2000a discusses the crevice corrosion resistance of zirconium in various chemical solutions and in Section 4.1.3 summarizes seven (7) crevice corrosion tests and reports that crevice corrosion was not observed. The U-bend tests discussed in Section 4.1.4 are also designed to produce crevice corrosion testing under the U-bend test washers. In these tests, no crevice corrosion is reported. Section 6.1.10 discusses the theoretical reasons why zirconium is immune to this type of corrosion. Yau and Webster 1987 ( p. 717) report that zirconium is among the structural metals that are most resistant to crevice corrosion. Greene et al. 2000 ( p. 7) also reported no ANI-WIS-MD-000008 REV 00 36 Apr112000 crevice corrosion. In summary crevice corrosion has not been observed in zirconium and is not expected at YMP conditions. TSPA Disposition This FEP is addressed in CRWMS M&O 2000a IRSR Issues: CLSTI, CLST3, TSPAI3, TSPAI4 References: Yau and Webster 1987 ,CRWMS M&O 2000a ,Greene et al. 2000, CRWMS M&O 2000a Basis for Screening Decision: Excluded based on screening argument presented above. 6.2.8 High Dissolved Silica Content of Waters Enhances Corrosion of Cladding - YMP No. 2.1.02.18.00 FEP Description It must be determined if the high dissolved silica content of waters enhances corrosion of cladding. Screening Decision: Exclude, low consequence Screening Argument: Silica is very stable. It is practically insoluble in water or acids, except hydrofluoric acid. It is not corrosive to most materials. However, there may be some fluoride contamination in silica, in which case it becomes corrosive to some materials, including zirconium, because fluorides in silica are soluble in water. For information on fluoride see YMP No. 2.1.02.16.00, Section 6.2.6. TSPA Disposition Silica content is not an issue, but the fluoride content that is in the silica is an issue. The fluoride issue is discussed in YMP No. 2.1.02.27.00. Therefore this FEP is excluded from further evaluation based on the arguments described below. IRSR Issues: CLST1, CLST3, TSPAI3, TSPAI4 References: CRWMS M&O 2000a, Yau and Webster 1987 Basis for Screening Decision Silica content is not an issue, but the fluoride content that is in the silica is an issue. Although frequently used as a corrosion resistant coating on materials, silicon dioxide reacts with hydrofluoric acid to form fluosilicic acid. Since fluosilicic acid is highly corrosive to zirconium even at room temperature, silicon dioxide can not be used as a technique to prevent attack of zirconium by acidic fluorides. The fluoride corrosion itself is addressed in another FEP ( YMP No. 2.1.02.27.00) and is included. The potential for silica itself degrading the cladding is negligible. Hansson (1984) reports corrosion tests with concrete pore fluids which normally contain silica. Yau (1983) reports corrosion tests in sea water which also contains silicon. Neither experimenters report significant corrosion. Both CRWMS M&O 2000a and Yau and Webster (1987, pp. Table 6) review the corrosion potentials for zirconium and show that it is resistant to most chemicals. ANI-WIS-MD-000008 REV 00 6.2.9 Creep Rupture of Cladding - YMP No. 2.1.02.19.00 FEP Description At high temperatures, creep rupture of Zircaloy cladding on spent fuel can occur and produce small perforations in the cladding to relieve stress. After the WP fails, the fuel can react with water and radioisotopes can thereby escape over time from the fuel rod. Screening Decision: Include. Screening Argument: Only moderately high temperatures (<350°C) are anticipated on the cladding surface and thus, creep rupture would not be a failure mechanism under storage conditions. Failure of Zircaloy cladding by creep rupture is included in the CSNF Cladding Degradation Component of the Waste Form Degradation Model as a perforation mechanism. Inclusion of the cladding creep process permits the TSPA-SR to evaluate, statistically, the potential for a small number of cladding ruptures because of a wide distribution of temperatures that may occur in the repository and a distribution of initial stresses that may be present in the spent fuel rods. Inclusion also permits an objective assessment of the cladding creep failures during storage and transportation prior to receipt at the repositoty. TSPA Disposition The manner in which creep rupture is discussed in detail in CRWMS M&O (2000~). In summary, to estimate the percent of the rods that will perforate from creep rupture, a creep failure criterion based on data from Chung et al. (1987, pp. 780-781) was used. Chung et al. (1987) conducted a series of 20 pressure burst tests at slow strain rates and mandrel tests with irradiated cladding. Four tests were on BWR Zircaloy 2 cladding with a burnup of 22 MWdIkgU and the remaining 16 tests were on PWR Zircaloy 4 cladding with a burnup of 28 MWdIkgU. For 18 of the 20 tests, strains when perforation occurred averaged 3.3% with a range of 0.4% to 11.7%. They inspected the perforations with a scanning electron microscope and found evidence that, in 11 of the 20 tests, the failures occurred at cracks formed in the cladding. Hence, the use of the Chung et al. (1987) results to estimate a failure criterion indirectly addresses the potential for lower failure strains from pre-existing internal cracks and the situation that perforation might not be caused by pure material creep. While the strain rates used in the experiments are faster than expected under repository conditions, they are not rapid burst tests and therefore are more applicable to repository conditions. The strain rates in Chung's gas pressurization experiments were such that the tests lasted between 1 and 312 hours with an average of 152 hours. The duration of the mandrel tests varied between 0.3 to 691.8 hours, with an average of 236 hours. Use of the Chung et al. (1987) empirical data to estimate a failure criterion implicitly assumes that the strain failure criterion conservatively bounds a failure criterion at slower strain rates. This assumption is valid since the creep tests summarized in CRWMS M&O (2000d, pp. 11-1 to 11-14) extend up to 10,000 hours (-1.14 year) and show creep strains, without failure, in excess of the failure criterion being used. In the CSNF Cladding Degradation Component, the potential for creep rupture is expressed statistically (CRWMS M&O 2000~) For a sampling of twelve thousand fuel rods and an average WP peak surface temperature of 277OC, the center rod peak temperature was 30S°C and the rod failure distribution was: ANI-WIS-MD-000008 REV 00 Lower limit = 1.06 % Mean - - 3.39 % Upper limit = 24.8 % Most of the predicted creep failures occurred in the rods with the highest initial stresses and during dry storage and transportation to the repository when temperatures above 350°C were possible. Only when WP peak surface temperatures reached 300°C did rod perforations from creep rupture start to be influential while in the repository. The FEP Thermally-Induced Stress Changes in Waste and EBS (YMP No. 2.1.11.07.00) is included in the TSPA-SR in that the stress in the fuel rods is adjusted for the temperature at each time and rod location when analyzing both creep damage and the potential for SSC. IRSR Issues: CLST1, CLST3, TSPAI4 References: CRWMS M&O 2000c, CRWMS M&O 2000d, Chung et al., 1987 Basis for Screening Decision No further elaboration in addition to the Screening Argument is necessq. 6.2.10 Pressurization From Helium Production Causes Cladding Failure - YMP No. 2.1.02.20.00 FEP Description Increased pressure within the fuel rod due to the production of helium gas could contribute to cladding failure. Screening Decision: Include. Screening Argument: Both delayed hydride cracking (DHC) and strain failures are driven by the cladding stress which may be caused by the internal gas (including initial fill gas, fission product gases, and helium gas from alpha decay) pressure buildup. The gas pressure will slowly increase over time by the production of helium as a result of alpha decay. TSPA Disposition This FEP is addressed in CRWMS M&O (2000b, Section 6.3.4). Both delayed hydride cracking (DHC) and strain failures are driven by the cladding stress which may be caused by the internal gas (including initial fill gas, fission product gases, and helium gas from alpha decay) pressure buildup. The gas pressure will slowly increase over time by the production of helium as a result of alpha decay Manaktala (1993, Figure 3-4, p. 3-12) presents the helium pressure buildup for 100 "C as a function of time for a PWR fuel rod with 36 MWdIkgU burnup and an assumed 100% helium release from the fuel into the fuel rod gap. This figure was used to develop an equation for helium buildup in a fuel rod. In TSPA-SR, the creep analysis was performed including the helium buildup at 100 years, a time well after the peak thermal period. IRSR Issues: CLST1, CLST3, TSPAI4 ANI-WIS-MD-000008 REV 00 39 References: Manaktala, H.K. 1993, CRWMS M&O 2000b Basis for Screening Decision No further elaboration in addition to the Screening Argument is necessary 6.2.11 Stress Corrosion Cracking (SCC) of Cladding - YMP No. 2.1.02.21.00 FEP Description Stress corrosion cracking mechanisms can contribute to cladding failure. These mechanisms can operate both from the inside out from the action of fission products, or from the outside in from the actions of salts or other chemicals within the waste package. Screening Decision: Include. Screening Argument: SCC requires a susceptible material, an aggressive chemical environment, and high stress levels. Iodine-induced SCC requires an iodine concentration in the fuel-cladding gap greater than 5 x g/cm2 (Cunningham et al. 1987, pp. A.4, A.5). Below this threshold of free iodine concentration, Zircaloy cracking due to SCC has not been observed. In actual fuel rods, free iodine concentrations are expected to be negligible. However, over sufficiently long times at high stress and elevated temperatures the iodine may be present in sufficient quantities for SCC, such that, once cracking starts, there is sufficient time to propagate through the cladding (without considering crack velocities). As reported by Tasooji et al. (1984, p.600, their Figure 3), a rod with a stress above 180 MPa fails by SCC. Although few rods have such high stresses and the same rods that fail from SCC are also prone to fail from creep, the inclusion of SCC increases the mean failure from creep alone from 2.0% to 2.4% for dry storage and transportation (CRWMS M&O 2000~). The SCC model compares the existing stress with a SCC threshold stress and considers neither the crack size, chemical environment, nor the crack location. Therefore this model is applicable for Inside-Out SCC (FEP YMP No. 2.1.02.21.01) or Outside-In SCC (FEP YMP No. 2.1.02.21.02). TSPA Disposition Over sufficiently long times at high stress and elevated temperatures, and with a sufficient concentration of free iodine present, SCC may occur in a few fuel rods. Once cracking starts, there is sufficient time to propagate through the cladding. IRSR Issues: CLST1, CLST3, TSPAI4 References: Cunningham, M.E. ; Simonen, E.P.; Allemann, R.T.; Levy, I.S.; and Hazelton, R.F. 1987, Tasooji, A,; Einziger, R.E.; and Miller, A.K. 1984, CRWMS M&O 2000c Basis for Screening Decision No further elaboration in addition to the Screening Argument is necessary. ANI-WIS-MD-000008 REV 00 6.2.12 Hydride Embrittlement of Cladding - YMP No. 2.1.02.22.00 FEP Description Cladding contains hydrides after reactor operation. The cladding might also pick up more hydrogen from cladding general corrosion (wet oxidation) after the WP is breached. Hydrides might also form from U02 oxidation (after WP and cladding perforation). In addition, hydrides may dissolve in warmer areas of the cladding and migrate to cooler areas. Hydrogen can also move from places of low stress to places of high stresses, causing hydride reorientation or delayed hydride cracking (DHC). The buildup of hydrides can cause existing cracks to propagate by DHC or hydride embrittlement. Screening Decision: Exclude, low probability. Screening Argument: Hydrogen will be generated in the WP as the WP internals corrode. This hydrogen is not expected to be absorbed directly by the fuel cladding, because the H2 molecules are not expected to migrate through the high-density Zr02 layer on the fuel cladding. Available data on zirconium hydriding indicate that corrosion of waste package internals will not result in hydriding of fuel cladding. Although hydriding of Zircaloy cladding as a result of galvanic corrosion has been observed, Zircaloy cladding alloys resist galvanic corrosion when contacting carbon steel, and, if such a galvanic corrosion effect should occur, corrosion of the carbon steel would quickly break the electrical contact between the two materials. Failure of the cladding by hydride reorientation is unlikely. Stresses and temperatures are too low for hydride reorientation to occur and the cladding material will maintain sufficient strength even if hydride reorientation did occur such that failure would not be expected. Failure of the cladding by DHC is unlikely and has not been included in the abstraction for the TSPA-SR. Stresses (and stress intensity factors) are too low for crack propagation. Failure of the cladding by hydrogen embrittlement is unlikely. Hydrogen absorption from fuel cladding surface oxidation and WP corrosion will be negligible. Hydrogen migration will be limited at the temperatures expected during disposal. TSPA Disposition Exclude IRSR Issues: CLST1, CLST3, TSPAI4 References: Baker 1992, ASTM B811, ASTM B855, Clayton, J. C. 1984, CRWMS M&O 1997, Cunningham, M.E.; Simonen, E.P.; Allemann, R.T.; Levy, I.S.; and Hazelton, R.F. 1987, Dieter, G.E. 1961, Einziger, R.E. et. al., 1982, Einziger, R.E. and Kohli, R. 1984, Garzarolli et al. 1979, IAEA 1998, Kreyns, P.H.; Bourgeois, W.F.; White, C.J.; Charpentier, P.L.; Kammenzind, B.F.; and Franklin, D.G. 1996, Laming, D.D.; Beyer, C.E.; and Painter, C.L. 1997, Mahmood, S.T.; Farkas, D.M.; Adamson, R.B.; and Etoh, Y. 1998, McEachern, R.J. and Taylor, P. 1997, McMinn, A,; Darby, E.D.; and Schofield, J.S. 2000, Peehs, M. and Fleisch, J. 1986, Pescatore, C.; Cowgill, M.G.; and Sullivan, T.M. 1990, Puls, M.P. 1988, Reed-Hill, R.E. 1973, Rothman, A.J. 1984, Shi, S.Q. and Puls, M.P. 1994, Wasywich, K.M. and Frost, C.R. 1992, Yau, T. L., 1984, Yau, T.L. 1983, YMP 1998, Clayton 1989, CRWMS M&O ANI-WIS-MD-000008 REV 00 41 Apr112000 2000b, CRWMS M&O 2000g, Smith 1966, Huang 1995, Peehs 1998, Paurbaix 1974, Mardon et al. 1997, ASM 1992. Basis for Screening Decision Hydrogen embrittlement results in a generally reduced resistance to fracture. In Zircaloy, hydrogen embrittlement is normally caused by formation of microscopic flakes of zirconium hydride. Since the hydride flakes are quite brittle, a crack can propagate more readily by preferentially following the hydrides. Resistance to fracture (fracture toughness KIC) is a measure of resistance to crack propagation through the material. Fracture toughness is typically measured in terms of the critical stress intensity factor, that is, the value of the stress intensity factor that will cause growth of a crack. The stress intensity factor is proportional to the far-field stress times the square root of the crack length. Kreyns et al. (1996, p. 767, Figure 5) show that such hydrides could decrease the fracture toughness (KIc) from 42 MPa.m O 5 to 8 ~ ~ a . m ' ~ as the hydrogen content increases from zero to 4000 ppm. As shown in CRWMS M&O 2000b (Section 6.10.2) the maximum stress intensity (KI) for the statistical distribution of rods and crack sizes at 260°C is 2.7 MPa mn and therefore even with hydride concentrations of 4000 ppm, failure is not expected. In the limit, 100 percent hydride and no metal at all, the fracture toughness is about 1 ~ ~ a . m ' ~ . The outer surface of the cladding could be fairly brittle (hydrogen content greater than 800 ppm) but much of the cladding thickness has a reasonable toughness. 6.2.12.1 Hydride Embrittlement from Zirconium Corrosion - YMP No. 2.1.02.22.01 Additional hydrides could be formed by general corrosion of the cladding. WPs are assumed to fail at various times, permitting water or steam to enter the failed WP. Wet oxidation (or steam oxidation if the local temperature is above boiling) occurs, and approximately 17 percent (Lanning et al. 1997, Vol. 1, p. 8.4, Figure 8.2) of the hydrogen released from the water is absorbed by the fuel cladding in the failed WP. However, the fuel cladding picks up very little hydrogen because the corrosion rate is so slow. Even the hottest rods will pick up only small quantities of hydrogen from cladding surface general corrosion if the WP remains sealed for the first 100 years. 6.2.12.2 Hydride Embrittlement from WP Corrosion & H2 Absorption - YMP No. 2.1.02.22.02 Many investigators have considered the hydriding of zirconium alloys (for example, the papers cited in Clayton 1989, Tables 1 through 4). Many of these investigations have been straightforward measurements of the rate of hydriding under various conditions. However, at least one set of experiments directly determined the origin of hydrogen in the metal. IAEA 1998 (p. 92) discusses experiments in which zirconium-base alloys were oxidized in normal water (H20) with dissolved tritium gas (T2). This experiment is extremely sensitive. If even one part in lo6 of the hydrogen was from dissolved gas, the radioactivity of T2 would result in thousands of decays per second for one square centimeter of surface. This level of activity would be readily detected. IAEA 1998 (p. 92) gives the following discussion of the experiment: Oxidation studies using T2/H20 mixtures ... have shown that, during normal oxidation, no T2 enters the metal ... until the thermally-induced exchange reaction has progressed to the point where a measurable fraction of HTO has been formed. Thus, the hydrogen isotopes which enter the metal do so as an integral part of the reaction of the zirconium with water ANL-WIS-MD-000008 REV 00 42 April 2000 molecules, and not by reaction with any dissolved hydrogen in the water. Studies have shown that this situation persists ... until hydrogen over-pressures in the system of tens of MPa are present. Note that, according to the quotation given above, the hydrogen pressures required to cause hydriding are quite large. By comparison, the highest credible hydrogen pressure in a breached waste package is pure hydrogen gas at atmospheric pressure, or about 0.1 MPa. It is instructive to examine why high hydrogen pressures can cause hydriding. Even at 200 OC, the solubility of oxygen in zirconium is greater than 1% by weight, and the solubility of oxygen increases with increasing temperature (Baker 1992, p. 2-326). As a result, the oxide film on the surface of a piece of zirconium is normally not thermodynamically stable. In its discussion of experimental procedure, Smith (1966 p. 325) notes "Zirconium samples were first annealed at 700" C under vacuum (- mm Hg) to remove any oxide film. The film dissolved into the sample, leaving them a bright metallic color." It is clear that the oxide film can be damaged or even destroyed by heat treatment in a suitable environment. However, the film can be maintained if there is a supply of oxygen. Water will serve as a source of oxygen, because the electrochemical domain of stability for zirconium metal lies well below that of water (Pourbaix 1974, p. 226). The effect of the oxygen supply has been studied. Garzarolli et al. (1979, p. 64) state that: the effect of the composition of the gas atmosphere on the electrical properties of ZrO, corrosion films was measured ... . The results revealed a large decrease in the electric resistance when the atmosphere changed from oxidizing to non-oxidizing, indicating a drastic change of the morphology (passivity of the oxide film) ... the obvious implication of all available results is that massive hydriding can start when the availability of oxygen to continuously repair the protective oxide film falls below a critical value. The conditions for hydriding have been quantified. Clayton (1989, p. 270) presents the equation where p ~ , o and p ~ , are the pressures of Hz0 and HZ, respectively. For Equation 2 to be applicable, both pressures must be given in torr (millimeters of Hg head). (This fact is deduced as follows. First, Clayton (1989, Tables 1 through 4) indicates that pressures are measured in "mm". Second, pressures of "760 mm" (of Hg) (= atmospheric pressure) occur many times in these tables. Third, the values of "Critical P ~ , ~ / ~ H , " given in the tables are consistent with both . . values being in the same units.) ForpH, = 101 kPa = 760 tom, it is found that mzo (protective) = 0.24 kPa = 1.8 tom. Note that, by atmospheric standards, this p ~ , o corresponds to a dry gas. For comparison, the vapor pressure of water at 25 "C is about 3.2 kPa. For this small amount of humidity, corrosion of the waste package internals is not credible. However, if no corrosion occurs, no hydrogen is produced, so hydriding is impossible It should be noted that hydriding of zirconium by absorption of gas has been observed in the laboratory. For example, Smith 1966 (Table 3) gives data on hydrogen absorption. However, the environment for these experiments was hot, extremely pure hydrogen. Smith 1966 (Table 3) states that the temperatures for the hydrogen absorption experiments were 210 OC to 700 OC. Smith 1966 (p. 325) notes that "hydrogen was purified by passing it through a Deoxo unit, a bed of platinized asbestos (300" C), a tube of P2O5 and a liquid nitrogen trap." The evident intention ANI-WIS-MD-000008 REV 00 43 Apr112000 is to react any oxygen impurities, and absorb or condense any water vapor that is formed. Mer this treatment, vety little oxygen would have been available to maintain the oxide film. Such an environment is not relevant for a repository at Yucca Mountain because air, water vapor, or liquid water will be present and will maintain the protective oxide film. It is understood that corrosion of waste package internals will occur at temperatures below the range for which the Equation 2 was developed. Data on hydriding at lower temperatures were not available in the literature. In some respects, it can be argued that it is conservative to apply Equation 2 to a repositoty. Equation 2 was developed for coupons of Zircaloy-2. This alloy is more susceptible to hydriding than is Zircaloy-4 (Clayton 1989, Table 5), and coupons are more susceptible to hydriding than is tubing (Clayton 1989, Table 5). Therefore, Equation 2 should give conservative predictions of the susceptibility of spent fuel cladding to hydriding. In summary, hydrogen will be generated in the WP as the WP internals corrode. This hydrogen is not directly absorbed, because the H2 molecules do not migrate through the high-density Zr02 layer on the cladding. Hydrogen is introduced into the coolant in PWRs to reduce oxygen ions and reduce corrosion of components and yet it is not absorbed into the cladding. This degradation mode is excluded from TSPA-SR. 6.2.12.3 Hydride Embrittlement from Galvanic Corrosion of WP Contacting Cladding - YMP No. 2.1.02.22.03 In current waste package designs, the fuel assemblies will be in contact with carbon steel tubes (CRWMS M&O 1997). Since the fuel cladding and WP basket tubes are of different materials, there may be galvanic corrosion with hydrogen charging of the fuel assembly. This would occur if there were electrical contact between the assembly and the fuel basket tube, oxidation of iron from the carbon steel fuel basket tubes, and reduction of hydrogen ions on the fuel assembly cladding. In addition, it would be necessary that at least some of the reduced hydrogen be absorbed by the metal of the fuel assembly cladding. It should be noted that there is only a limited period during which galvanic corrosion could occur in a specific WP. The fuel basket tubes will be completely corroded away within tens to a few hundreds of years after individual WP failures (YMP 1998, p. c-17). By the time the fuel basket tubes are completely corroded away in a specific failed WP, a thick layer of corrosion products will electrically insulate the fuel assembly cladding. It is important to understand the conditions for galvanic hydrogen charging. These can be inferred from results given in Yau (1983, p. 26/10). Yau discusses a series of experiments in which zirconium alloy U-bend samples were exposed to boiling seawater for 365 days. Each sample was loaded by an uninsulated steel coupling. Three compositions were considered: Zr 702, Zr 704 with nickel, and Zr 704 without nickel. Each composition was tested in both unwelded and welded conditions. Of the six samples, only the welded sample of Zr 704 with nickel showed hydrogen pickup. ("Zr 702" and "Zr 704" are taken to be UNS R607024 and R6070, respectively. ASTM B55UB551M-97 (Table 1) gives composition limits for these materials; the maximum hydrogen content is 50 ppm.) Except for the welded sample of Zr 704 with nickel, the U-bend samples had hydrogen contents of 5 to 9 ppm at the end of the test. It is apparent that contact with a steel surface alone does not lead to galvanic hydrogen charging. ANI-WIS-MD-000008 REV 00 44 Apr112000 Although Yau (1983, p. 2612) does not give the compositions of the samples of Zr 704 with nickel and Zr 704 without nickel, it is instructive to compare the amount of nickel in various zirconium alloys. According to the minimum concentrations for other elements, the specification for UNS R60704 admits up to 1.3% nickel, and that for UNS R60702 admits up to 0.8% nickel (ASTM B551, Table 1). UNS R60802 (Zircaloy-2) contains 0.03% to 0.08% nickel, and UNS R60804 (Zircaloy-4) contains no more than 0.007% nickel (ASTM B811-90, Table 2). It is clear that the Zircaloys contain very small amounts of nickel, so hydriding as a result of contact with the steel basket tubes is not expected. It might be argued that, because of its nickel content, Zircaloy-2 cladding could be hydrided. However, Zircaloy-2 fuel cladding is used only for boiling water reactors, and the fuel cladding will be separated from the basket tubes by the fuel channels. Hydriding of zirconium alloys as a result of contact with dissimilar materials has been observed, so it is important to consider the conditions that promote hydriding. Clayton (1984, p. 578) lists a series of experiments in which Zircaloy-4 fasteners, clamped onto a sample of nickel-base alloy were hydrided by exposure to hot water with dissolved hydrogen. Clayton (1984, p. 573) discusses the mechanism: ... in mechanically attaching a Zircaloy fuel rod fastener to an Inconel support plate, relative motion occurred during assembly. The relative motion was sometimes sufficiently severe and the bearing stress sufficiently high to overcome the protective effects of the corrosion oxide film and graphite lubricant. Smearing and bonding of Inconel onto local regions of the mating Zircaloy contact surface occurred and provided the potential for accelerated hydriding. Inconel has a relatively high permeability for hydrogen ... and acts as a "window" for hydrogen entry into the Zircaloy. A similar but very localized effect was noted for Zircaloy-4 smeared with Inconel alloy (4.6% nickel - 16.7% chromium - 3.5% copper - 75.2% iron). Although hydriding was observed in the experiments discussed above, it does not follow that hydriding will occur under repository conditions. As is noted in Clayton (1984, p. 587), "Nickel alloy smearing and bonding to filmed Zircaloy, rather than just tight surface contact, is necessary for accelerated hydriding." Fuel assemblies will be lowered carefully into waste package baskets. The high contact pressures typical of fastener tightening are not expected. After emplacement, the fuel assembly will simply rest on the basket. Therefore, smearing and bonding should not occur. If it is nevertheless supposed that there is some bonding between the fuel assembly and the basket, the significance of the effect must be examined. It should be noted that accelerated hydriding is a transient, not a persistent, effect. Clayton (1984, p. 572) notes that the initial high accelerated hydrogen ingress rate was effectively shut off during exposure at 271 "C in about 25 days. Because of the extremely small size of contact spots, corrosion of the carbon steel would quickly break the electrical contact between the two materials. Since the time of contact will be small, the amount of hydrogen absorbed as a result of contact will also be small. This conclusion is supported by the discussion of U-bend hydriding above. ANI-WIS-MD-000008 REV 00 45 The data above on zirconium hydriding indicate that corrosion of waste package internals will not result in hydriding of fuel cladding. Although hydriding as a result of galvanic corrosion has been observed, cladding alloys resist such corrosion on contact with carbon steel, and, if such an effect should occur, corrosion of the carbon steel would quickly break the electrical contact between the materials. 6.2.12.4 Delayed Hydride Cracking (DHC) of Cladding - YMP No. 2.1.02.22.04 CRWMS M&O (2000g, Section 6.3) reviews DHC and CRWMS M&O (2000b, Section 6.10.2) analyzes DHC with the predicted crack size distribution and rod stress distribution expected at YMP. The general DHC process begins with precipitation of a hydride at the crack tip. The stress gradient at the crack tip provides a driving force for preferential hydride precipitation in the tensile stress field in front of the crack tip. If the crack tip stress is greater than the fracture stress of the brittle hydride, the hydride fractures and crack growth occurs. There are three basic requirements for DHC: (1) incipient defects or cracks; (2) presence of hydride at or near the crack tip; and (3) sufficient stress to propagate the crack. Incipient cracks and defects are present in the cladding as a result of the manufacturing process and from irradiation. Hydrogen is present as a result of Zircaloy alloy impurities and in-reactor corrosion. Cooling of the fuel cladding after reactor operations (irradiation) promotes precipitation of hydrides. The high stresses required to crack a hydride exist only in front of a deep crack. The critical stress intensity for DHC is much greater than cladding stress intensities expected during dry storage. Additionally, at the high stress intensities required for DHC, SCC is the dominant crack growth mechanism. During delayed hydride cracking, hydrides slowly form at a crack until the crack propagates through the hydride region at the crack tip and stops. This sequence repeats itself and the crack propagates slowly through the metal. The hydrides preferentially collect at the crack tip because tensile stress reduces solubility in that region. The critical stress intensity factor (KIH) is the minimum stress intensity that will permit any DHC, regardless of velocity (velocity approaches 0). For this analysis (CRWMS M&O 2000a), the stress intensities (KI) will be calculated and compared to the KIH. If KI > KIH, then the crack will start to propagate and, because of long repositoty times, it is assumed that failure will occur. DHC failure occurred in some zirconium coolant tubes in a Candu reactor where high temperature gradients caused excess hydride buildup in a specific location. The stress intensity factor, KI, is a measure of the increased stress at the tip of a crack. The stress intensity factor is proportional to the far-field stress times the square root of the crack length. For a sharp crack a limiting case, the stress intensity factor is (Reed-Hill 1973, p. 800) where KI = Stress intensity factor, ~ ~ a . r n ' ~ o = Cladding stress, MPa w = Crack depth, m ANI-WIS-MD-000008 REV 00 Equation 3 is slightly modified from the form given by Reed-Hill; the crack depth w is used in place of c/2, where c is the crack length (Dieter 1961, p. 194). DHC is analyzed in CRWMS M&O (2000b, Section 6.10.2). The calculated crack size distribution is given in CRWMS M&O (2000b, Section 6.6 Figure 17). The median (P = 50 percent) value is 13 pm and the average crack is 18.6 pm. The largest size crack of the 2000 samples is 119 pm. The calculated stress distribution (CRWMS M&O 2000b, Section 6.7) is given in Figures 18 and 26 of that AMR. DHC is unlikely at temperatures above 260°C (Mahmood et al. 1998, p.20), because of the plasticity of the material. Rothman (1984, p. 37) reports that DHC is unlikely above 250°C because of the plasticity of the material. For this calculation, the temperature of 260°C is used and the pressure is adjusted accordingly. The crack size distribution and stress distribution can be combined to give the distribution of stress intensity factors, KT, in Figure 27 of that AMR. Some of the properties of this distribution of the stress intensity factor are: Mean: 0.47 M~a-m'.~ Median: 0.40 MPa-mo5 maximum: 2.7 MPa-mo5 minimum: 1.61E-03 MPa-mo5 The work of Shi and Puls (1994, p. 239, Fig. 7), shows experimental KIH in the range of 5 to 12 MPa-mo for zirconium alloy containing 2.5% Nb. Rothman (1984, p. 37), reports a KIH of 6 MPa-mo for Zircaloy-2. Pescatore et al. (1990, Table 6, p.50) report values of 5 and 14. Huang (1995, p. 195) shows KIH for irradiated Zircaloy-2 approaching 6 MPa-mo 5. For this AMR Huang's and Rothman's value for irradiated cladding of 6 MPa-mo was used. Because the observed values of K I are well below these values, exactly which value of K IH is selected is not important. No KI values in this AMR's sampling of rods are near the threshold stress intensity value. The maximum observed KI was 2.7 MPa-mo and the mean value was 0.47 MPa-mo 5. Rothman (1984, pp. 33 - 39) reviewed DHC in Zircaloy cladding in a repository. Rothman concludes that DHC is unlikely unless the fuel rods have large existing cracks (exceeding approximately 50 percent of wall thickness) and very high stresses (exceeding approximately 137 MPa). He also concludes that hydride reorientation is also unlikely because of the lack of large temperature gradients in the repository and the cladding stresses are lower than needed for reorientation. Peehs (1998, pp. 5, 6) concluded that neither DHC nor hydride reorientation would occur in dry storage. In conclusion, failure of the cladding by DHC in dry storage is unlikely and has not been included in the abstraction for the TSPA-SR analysis. Stresses (and stress intensity factors) are too low for crack propagation. Since the threshold for embrittlement failure is always greater than the threshold for DHC ( K I ~ > KIH), embrittlement failures are also not expected. ANI-WIS-MD-000008 REV 00 47 Apr112000 6.2.12.5 Hydride Reorientation - YMP No. 2.1.02.22.05 CRWMS M&O (2000g, Section 6.4) reviews hydride reorientation. The total hydrogen content in the fuel cladding is not as important as the amount of hydrogen in hydrides aligned perpendicular to the largest principal tensile stress. Normally, the hydrides are oriented in the circumferential direction and do not weaken the cladding against a hoop stress. Reorientation usually occurs under tensile stresses ranging from 69 to 208 MPa. Test rods at the low end of this range, but significantly higher than the stress expected at the repository temperatures, showed no reorientation of hydrides. This lack of reorientation indicates that there should be little, if any, cladding degradation due to hydrides under normal repository temperatures (Einziger et al. 1982, p. 65). In commercial reactor cladding during irradiation, hydrides form in a circumferential orientation (the normal to the platelet is in the radial direction). Such hydrides do not significantly weaken the cladding against hoop stress. In one dry storage test, reorientation to the radial direction was observed in one rod (Einziger and Kohli 1984, p. 119) although rod failure did not occur. This occurred in a fuel rod that had very high stresses (145 MPa at 323'C). Reorientation was not observed in fuel rods with lower stresses (13 to 26 MPa). The hydride reoriented so that the normals in the circumferential direction could possibly weaken the cladding. Reorientation was also observed in Candu reactor coolant pipes. Hydride reorientation under repository conditions was investigated as a potential cladding degradation mode. Figure 1 summarizes the data that Pescatore et al. (1990, pp. 52-55) collected showing reorientation. The points from Marshall show different fractions of reorientation. Hardie's point was reported at minimum stress. The straight solid line suggests a region in temperature and stress where reorientation occurs. This figure shows that at the cladding maximum temperature of 35OoC, the stress would have to be greater than 120 MPa before reorientation would start to occur. This corresponds to about 60 MPa at room temperature. From CRWMS M&O 2000b (Figure 26), it is estimated that 5% of the rods could have stresses this high and might undergo some reorientation. Rothman (1984) also studied cladding degradation in a repository and concluded that hydride reorientation would not occur. He was not considering fuel with burnups and stresses as high as considered in CRWMS M&O 2000b. The predicted cladding temperatures will be less than 350°C and therefore this comparison should be considered an upper bounding case. Pescatore et al. (1990, pp. 54, 69) state that even with hydride reorientation, stress levels will be insufficient to result in DHC and clad failure. For general DHC, they also said that crack propagation would tend to stop when the crack propagates into a hydride platelet. Puls (1988, p. 1507-1522) performed a series of strain tests on Zircaloy-2 with reoriented hydrides. His results are summarized in Table 3. He took samples of Candu coolant tubing and performed strain tests in the circumferential direction. The initial tubing is made in a similar fashion as cladding and develops hydride platelets with their normals in the radial direction. All tests were performed at room temperature. He used samples with both 20 ppm and 90 ppm hydrogen content and used two reorientation techniques to form hydrides of various lengths. One technique cooled the samples from 250°C with the stress near the yield point (designated y in Table 3). Other samples were cooled from 350°C at a stress of 200 MPa. This stress is more than twice than expected for ANI-WIS-MD-000008 REV 00 380 360 3 340 " 320 300 + $ 280 260 240 Hardie minimum o 220 200 120 140 160 180 200 220 Stress (M Pa) Figure 1. Temperature and Stress at which Hydride Reorientation was Observed for Cold Worked Zircaloy-2, 4 (Pescatore et al. 1990) most cladding in the repository (CRWMS M&O 2000b, Figure 26). Longer hydrides were produced by cooling the samples in a furnace, while shorter ones were produced by bench cooling the samples. Table 3 gives the range of hydride lengths. Two types of samples were used, smooth ones and notched ones (designated by n in the table). The table gives the stress (oy) for which 0.2 percent strain was measured. Also given is the ultimate stress at which necking and imminent failure was observed (or, for the arrested tests, was expected). For some experiments, the tests were stopped when the sample started to neck, but before failure (designated a for arrested in the table). In all of these tests, the reoriented hydrides did not significantly change the stress for 0.2percent strain or the ultimate stress. Both stresses are much higher than those expected in repository cladding. The yield and ultimate strains reported are also higher than the strain failure criteria of 1 percent used in the cladding strain failure analysis developed by Peehs and Fleisch (1986). AM-WIS-MDOOOOO8 REV 00 April 2000 Table 3. Puls' Zircaloy-2 Strain Tests on Zirconium with Reoriented Hydrides A It - t - -lype, y= nyariae reorienrea near yiela srress, n - norcnea, a = arresred (resr rermmarea oerore failure) i Source: M.P. Puls (1988, Tables 1,3,6) JU-JU I IOU '1. 11 13.0 Hydride reorientation might require all the hydride platelets to be dissolved before cool-down 50-90 50-90 50-90 50.90 50-90 and reprecipitation starts. This could be necessary because the hydrides prefer to condense at existing platelets and these earlier, existing platelets are oriented in the circumferential direction. For all the hydrides to dissolve at a maximum cladding temperature of 35OoC, the initial concentration must be less than 120 ppm (see Table 4). CRWMS (M&O 2000b, Figure 15) shows that only about 13 percent of the fuel has average concentrations this low. Knowing that the hydride concentration is directly proportional to the oxide thickness, CRWMS M&O (2000b, Figure 12) shows the fuel that could reorient under this hypothesis is the fuel with the lower burnups and therefore, lower stresses. Mardon et al. 1997 (their Figure 3, p. 408) shows hydride content as a function of burnup. To have less than 120 ppm, the figure shows that Zircaloy 4 fuel with burnups less than 22 MWdIkgU has low enough hydride concentrations to dissolve all the hydrides at 350°C. Again, these fuels would have the lowest stresses. The rod that Einziger observed reorientation was exposed to a 570°C transient which would support a solubility of 729 ppm (Table 4). It is possible that all the hydrides were dissolved in that experiment. In summary, failure of the cladding by hydride reorientation is unlikely and has not been included in the abstraction for TSPA-SR. Stresses and temperatures are too low for hydride reorientation to occur in most of the fuel and the cladding material will maintain sufficient toughness even if hydride reorientation did occur such that failure would not be expected. - 1 Y n n. Y n AM-WIS-MDOOOOO8 REV 00 50 April 2000 689 625 721 923 81 1 741 647 803 1032 936 4.1113.6 4.1113.6 Table4. Saturation Limits for Hydrogen in Zirconium as a Function of Temperature Temperature ("C) I H Concentration (ppm) I Arl I n I I Temperature ("C) 300 310 320 330 - I I Source: Pescatore et al. (1990. eq. 6) H Concentration (ppm) 66 75 85 96 - - I 44u I LOO 6.2.12.6 Hydride Axial Migration - YMP No. 2.1.02.22.06 Hydrides can form in cooler parts of the rod (end sections) because the hydrogen can dissolve into the fuel cladding metal matrix at a warmer area, diffuse toward the cooler area, and condense there. The effect was studied for dry storage (Cunningham et al. 1987, Appendix C) and for a 90-year period was determined not to be a problem. As the repository cools, the driving force for this redistribution (hydrogen solubility, temperature gradient, and diffusion rate) all decrease. The WP internals will act to minimize the temperature variation along the length of the fuel assembly. McMinn et al. (2000, Figure 15) show that a minimum of 42°C temperature difference between the hot location and cold location in the cladding is needed to move the hydrogen. At temperatures below 20OoC, temperature differences over 75°C are needed to move the hydrogen. It is unlikely that sufficient hydrogen can be moved because of a lack of large temperature gradients in the WPs. E E 450 6.2.12.7 Hydride Embrittlement from Fuel Reaction - YMP No. 2.1.02.22.07 Hydrides can be formed in the cladding from the oxidation of UOz fuel (Wasywich and Frost 1992, p. 1171). The fact that the fuel is oxidizing means that the cladding has already been perforated. This type of cladding degradation is only observed in BWRs and a high temperature steam environment is required for failure propagation. Such conditions are unlikely at YMP 31 3 ANI-WIS-MD-000008 REV 00 51 Apr112000 I 460 470 480 490 500 510 520 - 339 367 396 427 459 493 529 - - after WP failure. If the embrittlement becomes severe and the cladding is mechanically loaded in excess of its fracture toughness, the cladding could fail exposing additional fuel. This type of failure has a small effect on the fuel oxidation rate since the fuel oxidizes after the first penetration. It does affect the amounts and rate of fission product migrating out of the secondary phases of the fuel dissolution. This type of secondary cladding failure has not been modeled. 6.2.13 Cladding Unzipping - YMP No. 2.1.02.23.00 FEP Description In either dry or wet oxidizing conditions and with perforated fuel cladding, the U02 fuel can oxidize. The volume increase of the fuel as it oxidizes can create stresses in the cladding that may cause gross rupture of the fuel cladding (unzipping). Screening Decision: Dry Oxidation - Exclude, low probability. Wet Oxidation Include. Screening Argument: In the TSPA-SR model, wet unzipping is included as the key element in exposing and dissolving fuel from damaged cladding. Dry unzipping has been excluded because it is not expected to occur if the WP remains sealed for a few hundred years after repository closure and, if it were to occur, only a very small fraction of the fuel would undergo dry unzipping. TSPA Disposition Dry Oxidation - Exclude, low probability. Wet Oxidation - included as described in CRWMS M&O 2000e, and abstracted into TSPA-SR in CRWMS M&O 2000c. IRSR Issuance: CLST1, CLST3, TSPAI4 References: Doubt, G. 1984, Einziger, R.E.; Atkin, S.D.; Stellrecht, D.E.; and Pasupathi, V. 1982, Gray, W.J. and Wilson, C.N. 1995, Hanson, B.D. 1998, IAEA 1988, Johnson, A.B., Jr. 1977, Johnson, L.H. and Taylor, P.; 1998, McEachern, R.J. and Taylor, P. 1997, CRWMS M&O 1998, CRWMS M&O 1999c, CRWMS M&O 2000h, CRWMS M&O 2000e Basis for Screening Decision 6.2.13.1 Dry Oxidation of Fuel - YMP No. 2.1.02.23.02 The cladding that has failed during reactor operation, from creep rupture, or from DHC generally has small cladding perforations and very small areas of fuel exposed to the near-field environment. When the WP fails, the inert atmosphere inside the WP is replaced with the air and steam from the near-field environment. After WP failure, the near-field environment is almost completely steam and dry oxidation is not expected. If the dry oxidizing conditions existed after WP failure and with perforated cladding, the UOZ phase in the spent fuel can oxidize in two stages, first to U409 and then to U3O8, causing an increase of the spent fuel matrix volume. These are nominal formulas only and do not necessarily describe the actual stoichiometry of these oxides. The stress from the volume increase can unzip the clad, causing a gross rupture of ANI-WIS-MD-000008 REV 00 52 Apr112000 the clad and exposing the fuel inside. Oxidation also produces an increase in surface area and dissolution rate, the combined factor being 150 (Gray and Wilson 1995, p. vii). Dry fuel oxidation has been experimentally studied by Einziger et al. (1982, p. 65) and Hanson (1998, p. 2.3). McEachern and Taylor (1997, p. i) summarize the work of others. An improved model for dry unzipping is presented in (CRWMS M&O 2000h). For defective WPs (assumed to have failed at emplacement), it may be possible to unzip the cladding of the fuel rods that are perforated. This will probably not actually occur because the environment is mostly steam and the 0 2 partial pressure is low at this early time. Therefore, dry oxidation is unlikely. However if it is assumed that dry oxidation does occur at this early time, the degree of unzipping for defective WPs can be estimated. The combination of defective WPs, which is taken as 0.02 percent (Doubt 1984, p. 30) with premature rod failure, which is 1.25 percent including the stainless steel cladding, gives the fraction of rods expected to be converted to U3O8 to be 2.5 x (= 2 x l o 4 x 1.25 x lo-'). Overall, when considering the presence of steam, timing of WP failures and decreasing temperatures, it is expected that only very small fractions of fuel will be converted to U308, and these small fractions will not affect the PA. Earlier analysis presented in the TSPA-VA (CRWMS M&O 1998, Section 6.3.1.1.9) showed that the cladding could unzip if both the cladding is perforated and the WP also fails within the first 200 years. Such early failures of the WP are not expected. 6.2.13.2 Wet Oxidation of Fuel - YMP No. 2.1.02.23.03 Cladding could unzip in a wet environment. As the U02 is converted to secondary phases such as metaschoepite and Na-boltwoodite, the fuel volume increases. These phases will only form where there is available liquid volume, or, alternatively, they might form at solid surfaces and tear the cladding. Cladding unzipping has not been observed in spent fuel pools where fuel has been stored for tens of years as noted by the IAEA (1988, Table XXVI) and Johnson (1977, p. 20). The Canadians tested a defective rod in a steam autoclave for ten years and did not measure any cladding strain (Johnson and Taylor 1998, p. 10). While wet unzipping has not been observed, the time periods are very short compared to the repository scale and a form of wet oxidation and secondary phase formation might tear the cladding. The current wet unzipping model is developed in CRWMS M&O 2000e and the abstraction is described in CRWMS M&O 2000c (Section 6.6). All failed Zircaloy clad rods, all stainless steel clad rods (assumed to be failed) are available to unzip. The unzipping velocity is dependent on the fuel intrinsic dissolution rate. This rate is evaluated at each TSPA-SR time step because it depends on the temperature and chemistry inside the WP. 6.2.13.3 Volume Increase of Corrosion Products - YMP No. 2.1.09.03.00 The volume increase of corrosion products is included in the TSPA-SR. It is this volume increase from the corrosion of U02 to other minerals that drives the cladding wet unzipping described above. 6.2.13.4 Gap and Grain release of Cs, I - YMP No. 2.1.02.07.00 The release of fission products from the gap and grain boundaries is included in the TSPA-SR and described in CRWMS M&O 2000c (Section 6.5) ANI-WIS-MD-000008 REV 00 6.2.14 Mechanical Failure (of Cladding) - YMP No. 2.1.02.24.00 FEP Description Mechanical failure of cladding may result from external stresses, such as ground motion and rockfall during earthquakes. Screening Decision: Include. Screening Argument: The analysis of seismic events is included in the TSPA-SR. Seismic events with a frequency of 1.1~10.~ eventslyear are assumed to break all of the fuel cladding. Such events are sampled, and, when such an event occurs, all cladding is failed and is considered to be available for unzipping. The effect of a rubble bed consisting of rocks from a drift collapse on bare fuel rods (no WP or possible drip shield protection) is not included in TSPA-SR because this would not occur until well after the 10,000 years considered for the TSPA-SR. Since high accelerations are needed to fail the cladding, rock drops onto an intact WP will not cause rod failure and therefore was not included in the TSPA-SR. TSPA Disposition Mechanical failure will be represented in the performance assessment for the TSPA-SR. IRSR Issues: CLST1, CLST3, TSPAI4 References: CRWMS M&O 1999b, Witte et al. 1989 Basis for Screening Decision Seismic analysis (CRWMS M&O 1999b, Section 6.1) shows that most of the rods in the WPs would fail from a very severe earthquake (about once per million years event) but no rods would fail for less severe and moderate frequency seismic events. This is consistent with studies (Witte et al. 1989, p. 194) of rod damage during transportation accidents that concluded that 63 g accelerations are needed to fail the rods in the shipping container (or WP). Therefore, the seismic failures have been included in the TSPA-SR as a disruptive event. The analysis of seismic events is included in the TSPA-SR. Based on the analysis (CRWMS M&O 1999b, Section 6.1), seismic events with a frequency of 1 . 1 ~ 1 0 - ~ eventslyear would break most of the fuel. Such events are sampled, and, when such an event occurs, all cladding is failed and is considered to be available for unzipping. 6.2.14.1 Rockfall - YMP No. 2.1.07.01.00 CRWMS M&O 1999b (Section 6.2) also considered the effect of a rubble bed consisting of rocks from a drift collapse on bare fuel rods (no WP or possible drip shield protection). The analysis showed that the bare fuel assemblies would fail under the static loading of the rocks. This effect was not included in the cladding degradation abstraction because it does not occur until after the WP no longer affords any protection. The first patches (about 100 cm by 100 cm) penetrate the WP in 50,000 to 60,000 years, and a significant number of patches (about 100 patches) are open in about 350,000 years. With the potential for rubble bed damage not occuring until well after the 10,000 years considered for the TSPA-SR, rubble bed damage to the cladding ANI-WIS-MD-000008 REV 00 54 Apr112000 was neglected. Witte et al. 1989 (p. 194, Table 3) analyzed rod damage during transportation accidents that concluded that 63 g accelerations are needed to fail the rods in the shipping container (or WP). This high acceleration indicates that rock drops onto an intact WP will not cause rod failure and therefore this FEP was not included in the TSPA-SR. 6.2.15 Localized Corrosion Perforation from Fluoride - YMP No. 2.1.02.27.00 FEP Description Fluoride is present in Yucca Mountain groundwater, and zirconium has been observed to corrode in environments containing fluoride. Therefore, fluoride corrosion of cladding may occur in WPs. Screening Decision: Include. Screening Argument: Numerous processes have been identified that might cause localized corrosion and eventually perforation of Zircaloy cladding on CSNF. Yet as summarized in FEP 2.1.02.16.00 and discussed in more detail in the AMR, Clad Degvadation- Local Covvosion of Zivconium and Its Alloys Undev Reposito~) Conditions (ANL-EBS-MD- 000012) (CRWMS M&O 2000a), the conditions necessary for these processes generally do not exist in the repository. Even pitting corrosion in the presence of fluorides does not appear likely since the concentration of fluoride in pore water at Yucca Mountain is low, this low initial amount of fluoride is consumed during the corrosion process, and the pH of the in-package chemistry does not drop low enough to promote corrosion. Localized corrosion in the present of fluorides was included in the TSPA-SR to account for modeling uncertainty of the in-package chemistry since conditions for fluoride corrosion were considered more likely to occur relative to the other processes examined (CRWMS M&O 2000a). In the TSPA-SR, one important assumption was made to make the process feasible under the modeled conditions of the repository: the fluoride was assumed to remain at only one small segment of the cladding until the rod perforated, and not diffuse or migrate to other regions of the waste package (CRWMS M&O 2000~). TSPA Disposition Corrosion of zirconium has been observed in fluoride-containing environments. Since fluoride is present in Yucca Mountain groundwaters, localized corrosion from fluoride is considered as a perforation mechanism in the CSNF Cladding Degradation Component. Two scenarios for fluoride corrosion have been considered. In the first (water-filled WP scenario), the WP is full of water, and fluoride ions are transported to the cladding by aqueous diffusion. This scenario is less severe than the second one and is not addressed further. In the second (flow-through scenario), water enters the WP through a breach on the top and drips out through a breach on the bottom. The fluoride is assumed to be transported rapidly through the WP by advection, and fluoride attack is localized on a relatively small area of cladding (assumed to be 10 mm of rod length). It is also assumed that all the fluoride that enters the WP is consumed on the surface of the first rod it encounters until that rod fails. After one rod fails, the fluoride attacks the next rod. After WP breach, the fraction of cladding that fails is proportional to the volume of water that enters the package, reaching one (100%) when 2,424 m3 of water enters the WP. The uncertainty range placed on this estimate is plus or minus a factor of ten. In this approach, the fraction of cladding perforated is linearly dependent on the water inflow (percent failing equals 0.0413xm3 water in WP). The water flow into the WP increases with time as additional patches on the WP open. Cladding perforation rate also depends on the location of ANL-WIS-MD-000008 REV 00 55 April 2000 the WP group in the repository because of different drip rates in different repository regions (CRWMS M&O 2000~). IRSR Issues: CLST1, CLST3, TSPAI4 References: CRWMS M&O 2000a, CRWMS M&O 2000c Basis for Screening Decision No further elaboration in addition to the Screening Argument is necessary. 6.2.16 Diffusion-Controlled Cavity Growth (DCCG) - YMP No. 2.1.02.28.00 FEP Description Diffusion-Controlled Cavity Growth (DCCG) was once thought to be a possible creep rupture mechanism that could occur under the temperature and pressure conditions that prevailed during dry storage of spent fuel and might occur during disposal. Screening Decision: Exclude, low probability (general creep rupture process, however, is included-see FEP 2.1.02.19.00). Screening Argument: Applicants for dry storage licenses for CSNF were once required by the U.S. Nuclear Regulatory Commission (NRC) to assume diffusion controlled cavity growth (DCCG) to evaluate dry storage designs. That is, the NUREG-1536 (NRC 1997, Section 4.V.4.b) endorsement of the DCCG method which was used to calculate a maximum cladding temperature limit for a dry storage design. However, this design limit is overly restrictive and relatively inflexible. Recent literature does not support the use of this model for zirconiumbased materials (Pescatore and Cowgill, 1994, p. 83-85) since it has not been validated, and voids and cavities are rarely seen in irradiated Zircaloy. Pescatore and Cowgill (1994, p. 85) recommend a methodology similar to the approach used of calculating the amount of creep and comparing it to a creep failure criteria. The current NRC Interim Staff Guidance (ISG) Number 11 (NRC 2000) recognizes the controversy with the DCCG conceptual model and permits license applicants to use other creep models in their license application. The use of creep model presented in CRWMS M&O (2000c, Section 6.2) is consistent with this ISG. The German dry storage creep analysis approach (Peehs 1998) is also consistent with the methodology used in CRWMS M&O (2000c, Section 6.2) and cited in Section 6.2.9 of this AMR. DCCG is excluded as a creep rupture mechanism. Creep rupture, as a cladding perforation process, however, is included in the CSNF Cladding Degradation Component of the Waste Form Degradation Model as described in FEP 2.1.02.19.00. The CSNF Cladding Degradation Component uses a creep strain method based on phenomena expected to be encountered under dry storage conditions as explained in the AMR Clad Degradation-Summary and Abstraction (CRMWS M&O 2000c) TSPA Disposition Exclude. IRSR Issues: CLST1, CLST3, TSPAI4 ANI-WIS-MD-000008 REV 00 References: Pescatore, C. and Cowgill, M. 1994, NRC 1997, NRC 2000, CRWMS M&O 2000c Basis for Screening Decision No further elaboration in addition to the Screening Argument is necessary 6.3 NRC ISSUE RESOLUTION Both the expectations of and the strategic planning activities by the NRC emphasize the early identification and resolution of licensing issues, prior to the NRC's receipt of the license application to constmct a geologic repository. The NRC's objective is to reduce the number of issues and to better define in advance the issues that may be in dispute between licensee and regulator. NRC regulations and a 1993 agreement between NRC and DOE expand on this initiative by allowing NRC staff-level issue resolution to be achieved during the pre-licensing consultation period. Such resolution, however, would not preclude the issue's being raised and reconsidered during licensing proceedings. 6.3.1 Staff-Level Issue Resolution To structure staff-level interactions, NRC has focused on the topics the NRC considers most critical to post-closure performance of the proposed geologic repository. At present, NRC staff has developed 10 Key Technical Issues, nine of which relate to post-closure performance assessment. The nine KTIs are identified in Table 5. Table 5. Key Technical Issues Related to Post-Closure Performance Assessment Each KT1 is configured with sub-issues, sub-issue components, and the NRC's acceptance criteria, thus facilitating the NRC's issuance of Issue Resolution Status Reports. As an important part of the staff-level interaction process, the IRSR provides the primary mechanism that NRC staff uses to provide feedback to DOE. Number 1 2 3 4 Full resolution at the staff level concerning a particular issue is achieved during pre-licensing whenever the NRC staff have no further questions or comments regarding how the DOE program is addressing the issue, and the NRC has documented this situation. Furthermore, there may be some cases in which staff-level resolution during pre-licensing may be limited to documenting a common understanding regarding differences in NRC and DOE technical positions. Issue Total System Performance Assessment and Integration Container Lifetime and Source Term Evolution of the Near Field Radionuclide Transport Unsaturated and Saturated Flow Under Isothermal Conditions ANI-WIS-MD-000008 REV 00 57 Apr112000 o 1 Structural Defornrawrl aru ~ a ~ s r r l ~ t i ~ r y 9 I Igneous Activity 1 6.3.2 Cross-Reference Between FEPS Related to Cladding Degradation and NRC Key Technical Issues The following Tables 6 and 7 contain cross-references between NRC and Container Lifetime and Source Terms (CLST) and Total System Performance Assessment and Integration (TSPAI) Key Technical Issues and the cladding degradation FEPS. Table 6. Cross-Reference Between FEPS Related to Cladding Degradation and Container Life and Source Term Key Technical Issues rao on-c oes n ShF lconsoereo ne, processes for r i I ui I I uu lvvarenoggeo roos are reeaseo from me IShF oegraoal on an0 lne r e e a s l l 1 CLST Sub-Issue EBS through the oxidation and dissolution of sDent fl Acceptance Criteria FEPs 3. The rate at which 14. DOE has identifled and I reep rupture of cladding -essurization from He Droduction causes claddina of radionuclidesfrom the EBS, as follws: dissolution of the I. lradlated U02 matrix. with the consequent formation of secondary minerals and colloids: prompt release Of radionuclides: degradation in the dry air environment: degradation and failure of fuel cladding: preferential dissolution of intermetallics in DOE SNF: and release of radionuclides from the WP emplacement drifts. iffusion-Controlled Cavity Growth ockfall (Large Block) 2.1.02.13.00 2.1.02.14.00 2.1.02.15.00 2.1 O2I6OO 2.1 O2I7OO 2.1.02.18.00 2.1.02.19.00 21022000 ANI-WIS-MD-000008 REV 00 58 Apr112000 " " General corrosion of cladding Microbiologically Influenced Corrosion (MIC) of cladding Acid corosion of cladding from radiolysis Localized corrosion (pitting) of cladding Localized corrosion (crevice corosion) of cladding High dissolved silica content of waters enhances corosion of cladding - CI - PI Table 7. Cross-Reference Between FEPS Related to Cladding Degradation and Total System Performance Assessment and Integration Issues coiosion of cladding 2.1.02.19.00 Creep rupture of cladding 2.1 0220.00 Pressurization from He production causes cladding failure - 2.1.02.21.00 Si - 2.1.02.22.00 H' - 2.1.02.23.00 Cl - 2.1 0224.00 Mecnanlcal mure or claaalnq tress cotros on cracn ng SCC of c aoo ng )onoe emonn emenl of caoo ng 2 1 0- 01 00 I~UL1\1611 (LdlYC DlULR, 2 1 09 03 00 1 Jo -me ncrease of Corros on P r ~ a ~ c r s 2 1 11 --00 l ~ n e r m a .- no-cea Srress Cnanaes n Nasre ana 7. CONCLUSIONS The following conclusions may be drawn: b This AMR contributes to the Yucca Mountain scenario development methodology by screening the FEPs related to cladding degradation. b This AMR develops screening arguments and TSPA-SR disposition statements for these FEPs. This AMR provides information for the YMP FEP Database and guidance to TSPASR analyses, which are appropriate for both site recommendation and license application documents. 0 Screening decisions reached in this AMR are documented in Table 2. 0 By providing references to appropriate other AMRs, this AMR provides a valuable link between issues related to cladding degradation and the research directed at their resolution. 0 Finally, this AMR correlates FEPs related to cladding degradation with the NRC Container Life and Source Term and Total System Performance Assessment and Integration Key Technical Issues. This AMR uses no models and performs no analysis but only summarizes the models or analysis performed in the references. The references address cladding degradation before receiving it at YMP and after emplacement. Inherent is the assumption that the cladding is not damaged at the YMP surface facilities. This requires that the YMP will have both procedures and design ANI-WIS-MD-000008 REV 00 59 Apr112000 features for the surface facilities to ensure that cladding degradation does not occur at the surface facilities. This AMR was developed using the methodology recommended in ASTM C 1174. This document may be affected by technical product input information that requires confirmation. Any changes to the document that may occur as a result of completing the confirmation activities will be reflected in subsequent revisions. The status of the input information quality may be confirmed by review of the Document Input Reference System database. 8. INPUTS AND REFERENCES 8.1 Documents Cited Baker, H. 1992. "Binary Alloy Phase Diagrams." Volume 3 of ASM Handbook. Page 2.326. Materials Park Ohio: ASM International. TIC: 247622. Bradley, E.R.; Bailey, W.J.; Johnson, A.B., Jr.; and Lowry, L.M. 1981. Examination of Zircaloy- Clad Spent Fuel After Extended Pool Storage. PNL-3921. Richland, Washington: Pacific Northwest Laboratory. TIC: 230049. Chung, H.M.; Yaggee, F.L.; and Kassner, T.F. 1987. "Fracture Behavior and Microstructural Characteristics of Irradiated Zircaloy Cladding." Zirconium in the Nuclear Industry, Seventh International Symposium Sponsored by ASTM Committee B-10 on Reactive and Refractory Metals, Strasbourg, France, 24-27 June 1985. Adamson, R.B. and Van Swam, L.F.P., eds. Pages 775-801. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 238255. Clayton, J.C. 1984. "Out-of-Pile Nicklel Alloy-Induced Accelerated Hydriding of Zircaloy Fasteners." Zirconium in the Nuclear Industry: Sixth International Symposium, Vancouver, British Columbia, June 28- July 1, 1982. Franklin, D.G. and Adamson, R.B., eds. ASTM STP 824, 572-591. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 241417. Clayton, J.C. 1989. "Internal Hydriding in Irradiated Defected Zircaloy Fuel Rods." Zirconium in the Nuclear Industry: Eighth International Symposium, held 19-23 June 1988 at Sun Diego, Calfornia. ASTM STP 1023, 266-288. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 241414. Cragnolino, G. and Galvele, J.R. 1978. "Anodic Behavior and Pitting of Zirconium and Zircaloy- 4 in Aqueous Solutions of Sodium Chloride." Passivity of Metals, Proceedings of the Fourth International Symposium on Passivity. Frankenthal, R.P. and Kruger, J., eds. p. 1053- 1057. Princeton, New Jersey: Electrochemical Society. TIC: 237155. Cranwell, R.M.; Guzowski, R.V.; Campbell, J.E.; and Ortiz, N.R. 1990. RiskMethodology for Geologic Uisposal of Radioactive Waste Scenario Selection Procedure. N UKEWCK- 1667. Washington, D.C.: U.S. Nuclear Regulatory Commission. ACC: NNA.19900611.0073. ANI-WIS-MD-000008 REV 00 CRWMS M&O 1995a. Total System Performance Assessment - 1995: An Evaluation of the Potential Yucca Mountain Repository. B00000000-01717-2200-00136 REV 01. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19960724.0188. CRWMS M&O 1995b. Analysis of Degradation Due to Water and Gases In MPC. BB0000000- 01717-0200-00005 REV 01. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19960419.0202. CRWMS M&O 1997. 21-PWR Waste Package Disposal Container. BBA000000-01717-3300- 00003 REV 00. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19971201.0260. CRWMS M&O 1998. "Waste Form Degradation. Radionuclide Mobilization. and Transport Through the Engineered Barrier System." Chapter 6 of Total System Performance Assessment- Viability Assessment (TSPA-VA) Analyses Technical Basis Document. B00000000-01717-4301- 00006 REV 01. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19981008.0006. CRWMS M&O 1999a. Cladding FEPs Screening Arguments. Development Plan TDP-WISMD- 000012 REV 00. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19990902.0472. CRWMS M&O 1999b. Breakage of Commercial Spent Fuel Cladding by Mechanical Loading. CAL-EBS-MD-000001 REV 00. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19991213.0237. CRWMS M&O 1999c. Conduct ofPerformance Assessment. Activity Evaluation, September 30, 1999. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19991028.0092. CRWMS M&O 1999d. Dose Calculation for the Single-CRMZI-PWR Waste Package with Drip Shield. BBAC00000-01717-0210-00017 REV 00. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19990927.0471. CRWMS M&O 2000a. Clad Degradation - Local Corrosion of Zirconium and Its Alloys under Repository Conditions. ANL-EBS-MD-000012 REV 00. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.20000405.0479. CRWMS M&O 2000b. Initial Cladding Condition. ANL-EBS-MD-000048 REV 00. Las Vegas, Nevada: CRWMS M&O. Submit to RPC URN-0246. CRWMS M&O 2000c. Clad Degradation - Summary and Abstraction. ANL-WIS-MD-000007 REV 00. Las Vegas, Nevada: CRWMS M&O. Submit to RPC URN-0276. CRWMS M&O 2000d. Comparison of Creep Correlations. Input Transmittal PA-WP-00048.Ta. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.20000223.0002. CRWMS M&O 2000e. Clad Degradation - Wet Unzipping: Release Rates from Breached Cladding and Potential Unzipping Velocity. ANL-EBS-MD-000014 REV 00. Las Vegas, Nevada: CRWMS M&O. Submit to RPC URN-0247. ANI-WIS-MD-000008 REV 00 CRWMS M&O 2000f. In-Package Chemistry Abstraction. ANL-EBS-MD-000037 REV 00. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.20000418.0818. CRWMS M&O 2000g. Hydride-Related Degradation of SNF Cladding Under Repository Conditions. ANL-EBS-MD-000011 REV 00. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.20000319.0048. CRWMS M&O 2000h. Clad Degradation - Dry Unzipping. ANL-EBS-MD-000013 REV 00. Las Vegas, Nevada: CRWMS M&O. Submit to RPC URN-0275. Cunningham, M.E.; Simonen, E.P.; Allemann, R.T.; Levy, I.S.; Gilbert, E.R.; and Hazelton, R.F. 1987. Control of Degradation of Spent LWR Fuel During Dry Storage in an Inert Atmosphere. PNL-6364. Richland, Washington: Pacific Northwest Laboratory. TIC: 210249. Dieter, G.E. 1961. Mechanical Metallurgy. 1'" Edition. New York New York: McGraw-Hill Book Company. TIC: 245037. Doubt, G. 1984. Assessing Reliability and Useful Lfe of Containers for Disposal of Irradiated Fuel Waste. AECL8328. Chalk River, Ontario, Canada: Atomic Energy of Canada Limited. TIC: 227332. Dyer, J.R. 1999. "Revised Interim Guidance Pending Issuance of New U.S. Nuclear Regulatory Commission (NRC) Regulations (Revision 01, July 22, 1999), for Yucca Mountain, Nevada." Letter from Dr. J.R. Dyer (DOE/YMSCO) to Dr. D.R. Wilkins (CRWMS M&O), September 3, 1999, OL&RC:SB-1714, with enclosure, "Interim Guidance Pending Issuance of New NRC Regulations for Yucca Mountain (Revision Ol)." ACC: MOL.19990910.0079. Einziger, R.E.; Atkin, S.D.; Stellrecht, D.E.; and Pasupathi, V. 1982. "High Temperature Postirradiation Materials Performance of Spent Pressurized Water Reactor Fuel Rods Under Dry Storage Conditions." Nuclear Technology, 57, (I), 65-80. Hinsdale, Illinois: American Nuclear Society. TIC: 237142. Einziger, R.E. and Kohli, R. 1984. "Low-Temperature Rupture Behavior of Zircaloy-Clad Pressurized Water Reactor Spent Fuel Rods Under Dry Storage Conditions." Nuclear Technology, 67, (I), 107-123. Hinsdale, Illinois: American Nuclear Society. TIC: 216868. Einziger, R.E. 1994. "Preliminary Spent LWR Fuel Oxidation Source Term Model." High Level Radioactive Waste Management, Proceedings of the F$h Annual International Conference, Las Vegas, Nevada, May 22-26, 1994. 2, 554-559. La Grange Park, Illinois: American Nuclear Society. TIC: 210984. EPRI (Electric Power Research Institute) 1997. The Technical Basis for the Classi$cation of FailedFuel in the Back-End of the Fuel Cycle. EPRI TR-108237. Palo Alto, California: Electric Power Research Institute. TIC: 236839. ANI-WIS-MD-000008 REV 00 Garzarolli, F.; von Jan, R., and Stehle, H. 1979. "The Main Causes of Fuel Element Failure in Water-cooled Power Reactors." Atomic Energy Review, 17, 31-128. Vienna, Austria: International Atomic Energy Agency. TIC: 221365. Library Tracking Number - 221365. Gray, W.J. and Wilson, C.N. 1995. Spent Fuel Dissolution Studies FY 1991-1994. PNL-10540. Richland, Washington: Pacific Northwest Laboratory. ACC: MOL. 19980625.0338. Greene, C.A.; Brossia, C.S.; Dunn, D.S.; and Cragnolino, G.A. 2000. "Environmental and Electrochemical Factors on the Localized Corrosion of Zircaloy-4." Corrosion 2000. Paper No. 210, Pages 1-16. Houston, Texas: NACE International. TIC: 246988. Hanson, B.D. 1998. The Burnup Dependence of Light Water Reactor Spent Fuel Oxidation. PNNL-11929. Richland, Washington: Pacific Northwest National Laboratory. TIC: 238459. Hansson, C.M. 1984. The Corrosion of Zircaloy 2 in Anaerobic Synthetic Cement Pore Solution. SKB-KBS-TR-84-13. Stockholm, Sweden: Svensk Kiirnbransleforsorjning A.B. TIC: 206293. Hillner, E.; Franklin, D.G.; and Smee, J.D. 1998. The Corrosion of Zircaloy-Clad Fuel Assemblies in a Geologic Repository Environment. WAPD-T-3 173. West MiMin, Pennsylvania: Bettis Atomic Power Laboratory. TIC: 237127. Huang, F.H. 1995. Fracture Properties of Irradated Alloys. 418 p. Richland, Washington: Avante Press. TIC: 224548. IAEA (International Atomic Energy Agency) 1988. Survey of Experience with Dry Storage of Spent Nuclear Fuel and Update of Wet Storage Experience. Technical Report Series, No. 290. Vienna, Austria: International Atomic Energy Agency. TIC: 233246. IAEA (International Atomic Energy Agency) 1993. Corrosion of Zirconium Alloys in Nuclear Power Plants. IAEA-TECDOC-684. pp. 82-92. Vienna, Virginia: International Atomic Energy Agency. TIC: 246386. IAEA (International Atomic Energy Agency) 1998. "Generic Formulation for Semi-Empirical Models." Section 7.3.1 of Waterside Corrosion of Zirconium Alloys in Nuclear Power Plants. IAEA-TECDOC-996. p. 178. Vienna, Austria: International Atomic Energy Agency. TIC: 238879. Johnson, A.B., Jr. 1977. Behavior of Spent Nuclear Fuel in Water Pool Storage. BNWL-2256. Richland, Washington: Pacific Northwest Laboratory. TIC: 234703. Johnson, A.B., Jr.; Bailey, W.J.; Schreiber, R.E.; and Kustas, F.M. 1980. Annual Report - FY 1979. Spent Fuel andFuel Pool Component Integrity. PNL-3 171. Richland, Washington: Pacific Northwest Laboratory. TIC: 229623. ANI-WIS-MD-000008 REV 00 Johnson, A.B., Jr.; Dobbins, J.C.; Zaloudek F.R.; Gilbert, E.R.; and Levy, I.S. 1987. Assessment of the Integrity of Spent Fuel Assemblies Used 7n Dry Storage Demonstrations at the Nevada Testsite. PNL-6207. Richland, Washington: Pacific Northwest Laboratory. TIC: 230135. Johnson, L.H. and Taylor, P. 1998. Alteration of Spent CANDUFuel In Aerated Steam at 150°C. Draft. Pinawa, Manitoba, Canada: Atomic Energy of Canada Limited. ACC: MOL.19981118.0337. Knoll, R.W., Gilbert, E.R. 1987. Evaluation of Cover Gas Impurities and their Effects on the Dry Storage of LWR Spent Fuel. PNL-6365. Richland, Washington: Battelle Memorial Institute, Pacific Northwest Laboratory. TIC: 213789. Kohli, R. and Pasupathi, V. 1986. Investigation of Water-logged Spent Fuel Rods Under Dry Storage Conditions. PNL-5987. Richland, Washington: Pacific Northwest Laboratory. TIC: 246472. Krauskopf, K.B. and Bird, D.K. 1995. Introduction to Geochemistry. 3rd Edition. New York, New York: McGraw-Hill. TIC: 2393 16. Kreyns, P.H.; Bourgeous, W.F.; White, C.J.; Charpentier, P.L.; Kammenzind, B.F.; and Franklin, D.G. 1996. "Embrittlement of Reactor Core Materials." Zirconium in the Nuclear Industry, Eleventh International Symposium held in Garmisch-Partenkirchen, Germany, September 11-14, 1995. Bradley, E.R. and Sabol, G.P., eds. ASTM STP 1295, 758-782. West Conshohocken, Pennsylvania: American Society for Testing and Materials. TIC: 237256. Lanning, D.D.; Beyer, C.E.; and Painter, C.L. 1997. FRAPCON-3: Modifications to Fuel Rod Material Properties and Performance Models for High-Burnup Application. NUREGICR-6534. Volume 1. Richland, Washington: Pacific Northwest National Laboratory. TIC: 238923. Little B. and Wagner P. 1996. "An Overview of Microbiologically Influenced Corrosion of Metals and Alloys Used in the Storage of Nuclear Wastes." Canadian Journal ofMicrobiology, 42, (4), 367-374. Ottawa, Canada: National Research Council of Canada. TIC: 246614. Maguire, M. 1984. "The Pitting Susceptibility of Zirconium in Aqueous C1-, Br-, and ISolutions." Industrial Applications of Titanium and Zirconium: Third Conference, A Symposium Sponsored by ASTM Committee B-10 on Reactive and Refractory Metals and Alloys, New Orleans, Louisiana, September 21-23, 1982. ASTM STP 830, p. 177-189. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 237161. Mahmood, S.T.; Farkas, D.M.; Adamson, R.B.; and Etoh, Y. 1998. "Post-Irradiation Characterization of Ultra High Fluence Zircaloy-2 Plate." Zirconium in the Nuclear Industry, Twelfth International Symposium held in Toronto, Ontario, Canada, June 15-18, 1998. 19-20. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 237270. ANI-WIS-MD-000008 REV 00 Manaktala, H.K. 1993. Characteristics of Spent Nuclear Fuel and Cladding Relevant to High- Level Waste Source Term. CNWRA 93-006. NRC-02-88-005. San Antonio, Texas: Center for Nuclear Waste Regulatory Analyses. TIC: 208034. Mardon, J.P.; Garner, G.; Beslu, P.; Charquet, D.; and Senevat, J. 1997. "Update on the Development of Advanced Zirconium Alloys for PWR Fuel Rod Claddings." Proceedings of the 1997 International Topical Meeting on Light Water Reactor Fuel Performance, Portland, Oregon, March 2-6. 405-412. La Grange Park Illinois: American Nuclear Society, Inc. TIC: 232556. McDonald. S.G. and Kaiser. R.S. 1985. "The Impact of Metallic Debris on Fuel Performance - A Case History." Proceedings, American Nuclear Society Topical Meeting on Light Water Reactor Fuel Performance. Orlando, Florida, April 21-24, 1985. DOEmE/34130-1, Volume 1, 2-(3- 19). La Grange Park, Illinois: American Nuclear Society. TIC: 226810. McEachern, R.J. and Taylor, P. 1997. A Review of the Oxidation of Uranium Dioxide at Temperatures Below 400°C. AECL-11335. Pinawa, Manitoba, Canada: Atomic Energy of Canada Limited. TIC: 232575. McKinnon, M.A. and Doherty, A.L. 1997. Spent Nuclear Fuel Integrity During Dry Storage - Performance Tests and Demonstrations. PNNL-11576. Richland, Washington: Pacific Northwest Laboratory. TIC: 237126. McMinn, A,; Darby, E.C.; and Schofield, J.S. 2000. "The Terminal Solid Solubility of Hydrogen in Zirconium Alloys." Zirconium in the Nuclear Industry: Twelfth International Symposium, Toronto, Canada, 15-18 June 1998. Sabol, G.P. and Moan, G.D., eds. Pages 173-195. West Conshohocken, Pennsylvania: American Society for Testing and Materials. TIC: 247102. McNeil, M. and Odom, A. 1994. "Thermodynamic Prediction of Microbiologically Influenced Corrosion (MIC) by Sulfate-Reducing Bacteria (SRB)." Microbiologically Influenced Corrosion Testing. ASTM STP 1232. Pages 173-179. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 246989. Peehs, M. 1998. Assessment of Dry Storage Performance of Spent LWR Fuel Assemblies with Increasing Burn-Up. Erlangen, Germany: Bereich Energieerzeugung. TIC: 245171. Peehs, M. and Fleisch, J. 1986. "LWR Spent Fuel Storage Behaviour." Journal of Nuclear Materials, 137, (3), 190-202. Amsterdam, The Netherlands: North-Holland Publishing Company. TIC: 235595. Pescatore, C. and Cowgill, M. 1994. Temperature Limit Determination for the Inert Dry Storage of Spent Nuclear Fuel. EPRI TR-103949. Palo Alto, California: Electric Power Research Institute. TIC: 102933. ANI-WIS-MD-000008 REV 00 Pescatore, C.; Cowgill, M.G.; and Sullivan, T.M. 1990. Zircaloy Cladding Performance under Spent Fuel Disposal Conditions Progress ReportMay 1 - October 31, 1989. BNL 52235. Upton, New York: Brookhaven National Laboratory. ACC: NNA. 19900710.0055. Pourbaix, M. 1974. Atlas of Electrochemical Equilibria in Aqueous Solutions. Houston, Texas: National Association of Corrosion Engineers. TIC: 208955. Puls, M.P. 1988. "The Influence of Hydride Size and Matrix Strength on Fracture Initiation at Hydrides in Zirconium Alloys." Metallurgical Transactions A, 19A, (6), 1507-1522. Metals Park, Ohio: American Society for Metals. TIC: 237143. Ravier, G.; Masuy, G.; and Willse, J.T. 1997. "Framatome and FCF Recent Operating Experience and Advanced Features to Increase Performance and Reliability." Proceedings of the 1997 International Topical Meeting on Light Water Reactor Fuel Performance, Portland, Oregon, March 2-6, 1997. 31-36. La Grange Park, Illinois: American Nuclear Society. TIC: 232556. Reed-Hill, R.E. 1973. Fuel in a TuffRepository. Physical Metallurgy Principles. 2nd Edition, 797-801. New York New York: D. Van Nostrand Company. TIC: 237154. Rothman, A.J. 1984. Potential Corrosion and Degradation Mechanisms of Zircaloy Cladding on Spent Nuclear Fuel in a Tuff Repository. UCID-20172. Livermore, California: Lawrence Livermore National Laboratory. ACC: NNA. 19870903.0039. Sanders, T.L.; Seager, K.D.; Rashid, Y.R.; Barrett, P.R.; Malinauskas, A.P.; Einziger, R.E.; Jordan, H.; Duffey, T.A.; Sutherland, S.H.; and Reardon, P.C. 1992. A Method for Determining the Spent-Fuel Contribution to Transport Cask Containment Requirements. SAND90-2406. Albuquerque, New Mexico: Sandia National Laboratories. TIC: 232162. Sasaki, S. and Kuwabara, S. 1997. "Utility Perspective on Commercial Light-Water Reactor Fuel in Japan." 1997 International Topical Meeting on Light Water Reactor Fuel Performance, Portland, Oregon, March 2-6, 1997. 11-20. La Grange Park Illinois: American Nuclear Society. TIC: 232556. Schneider, K.J. and Mitchell, S. J. 1992. Foreign Experience on Effects ofExtended Dry Storage on the Integrity of Spent Nuclear Fuel. PNL-8072. Richland, Washington: Pacific Northwest Laboratory. TIC: 205836. Shi, S.Q. and Puls, M.P. 1994. "Criteria for Fracture Initiation at Hydrides in Zirconium Alloys I. Sharp Crack Tip." Journal of Nuclear Materials, 208, (3), 232-242. Amsterdam, The Netherlands: North-Holland Publishing Company. TIC: 237135. Smith, T. 1966. "Kinetics and Mechanism of Hydrogen Permeation of Oxide Films on Zirconium." Journal ofNuclear Materials, 18, 323-336. Amsterdam, The Netherlands: Elsevier Science Publishers. TIC: 247624. ANI-WIS-MD-000008 REV 00 Tasooji, A,; Einziger, R.E.; and Miller, A.K. 1984. "Modeling of Zircaloy Stress-Corrosion Cracking: Texture Effects and Dry Storage Spent Fuel Behavior." Zirconium in the Nuclear Industry, Sixth International Symposium, Vancouver, British Columbia, June 28- July 1, 1982. Franklin, D.G. and Adamson, R.B., eds. ASTM STP 824, 595-626. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 241417. Uziemblo, N.H. and Smith, H.D. 1989. An Investigation of the Influence of Fluoride on the Corrosion of Zircaloy-4: Initial Report. Draft. PNL-6859. Richland, Washington: Pacific Northwest Laboratory. ACC: MOL. 19980326.0369. Van Konynenburg. R.A.: Curtis. P.G.: and Summers. T.S.E. 1998. Scopng Corroszon Tests on Candidate Waste Package BasketMaterials for the Yucca Mountain Project. UCRL-ID-130386. Livermore, California: Lawrence Livermore National Laboratory. ACC: MOL. 19980727.0385. Wasywich, K.M. and Frost, C.R. 1992. "Behavior of Used CANDU Fuel Stored in 150 C Moisture - Saturated Air." High Level Radioactive Waste Management, Proceedings of the Third International Conference, April 12-1 6, 1992 Las Vegas, Nevada. 1, 1166-1 173. La Grange Park, Illinois: American Nuclear Society. TIC: 4447. Witte, M.C.; Chun, R.C.; and Schwartz, M.W. 1989. "Dynamic Impact Effects on Spent Fuel Assemblies." 9th International Symposium on the Packaging and Transportation of Radioactive Materials, Washington, D.C., June 11-16, 1989. 1, 186-194. Oak Ridge, Tennessee: Oak Ridge National Laboratory. TIC: 240741. Yang, R.L. 1997. "Meeting the Challenge of Managing Nuclear Fuel in a Competitive Environment." Proceedings of the 1997 International Topical Meeting on LWR Fuel Performance, Portland, Oregon, March 2-6, 1997. 3-10. La Grange Park, Illinois: American Nuclear Society. TIC: 232556. Yau, Te-Lin 1983. "Corrosion Properties of Zirconium in Chloride Solutions." Corrosion 83, International Corrosion Forum, Anaheim, California, April 18-22, 1983. Paper No. 26, Pages 2611 - 26/13. Houston, Texas: National Association of Engineers. TIC: 243849 Yau, T. L. 1984. "Zirconium Versus Corrosive Species in Geothermal Fluids." Paper 140 of Corrosion 84. Houston, Texas: National Association of Corrosion Engineers. TIC: 238987. Yau, T.L. and Webster, R.T. 1987. Corrosion of Zirconium and Hafnium. Volume 13 of A m Handbook (Formerly Ninth Edition, Metals Handboog. Pages 707-721. Materials Park Ohio: American Society for Metals. TIC: 240704. YMP (Yucca Mountain Project) 1998. Disposal Criticality Analysis Methodology Topical Report. YMPITR-004Q, Rev. 0. Las Vegas, Nevada: Yucca Mountain Site Characterization Office. ACC: MOL. 19990210.0236. ANI-WIS-MD-000008 REV 00 8.2 Codes, Standards, Regulations and Procedures 64 FR 46976. 40 CFR 197: Environmental Radiation Protection Standards for Yucca Mountain, Nevada; Proposed Rule. Readily Available 64 FR 8640. Disposal ofHigh-Level Radioactive Wastes in a Proposed Geologic Repository at YuccaMountain, Nevada. Proposed rule 10 CFR 63. Readily Available AP-2.13Q, Rev. 0, ICN 0. Technical Product Development Planning. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.19990701.0617. AP-3.10Q, Rev. 2, ICN 0. Analyses and Models. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20000217.0246. AP-SI.lQ, Rev. 2 ICN 4. SoftwareManagement. Washington, DC: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20000223.0508 ASTM B 551lB 551M-97. 1997. Standard Specij7cation for Zirconium and Zirconium Alloy Strip, Sheet, and Plate. West Conshohocken, Pennsylvania: American Society for Testing and Materials. TIC: 247045. ASTM B 81 1-90. 1991. StandardSpecijkation for Wrought Zirconium Alloy Seamless Tubes for Nuclear Reactor Fuel Cladding. Philadelphia, Pennsylvania: American Society for Testing and Materials. TIC: 239780. ASTM C 1174-97. 1997. Standard Practice for Prediction of the Long-Term Behavior of Materials, Including Waste Forms, Used in Engineered Barrier Systems JEBS) for Geological Disposal of High-level Radioactive Waste. West Conshohocken, Pennsylvania: American Society for Testing and Materials. TIC: 246015. DOE (U.S. Department of Energy) 2000. Quality Assurance Requirements and Description. DOEIRW-0333P, Rev. 9. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.19991028.0012. NLP-2-0, Rev. 5. Determination of Importance Evaluations. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19981116.0120. NRC (U.S. Nuclear Regulatory Commission) 1997. Standard Review Plan for Dry Cask Storage Systems: Final Report. NUREG-1536. Washington, D.C.: U.S. Nuclear Regulatory Commission. TIC: 232373. NRC (U.S. Nuclear Regulatory Commission) 1998. Issue Resolution Status Report Key Technical Issue: Total System Performance Assessment and Integration. Rev. 1. Washington, D.C.: U.S. Nuclear Regulatory Commission. ACC: MOL.19990105.0083. ANI-WIS-MD-000008 REV 00 NRC (U.S. Nuclear Regulatory Commission) 1999. Issue Resolution Status Report Key Technical Issue: Container Lfe and Source Term. Rev. 2. Washington, D.C.: U.S. Nuclear Regulatory Commission. TIC: 245538. NRC (U.S. Nuclear Regulatory Commission) 2000. Interim Staff Guidance-I1 - Storage of High Burnup Spent Fuel. Washington, D.C.: U.S. Nuclear Regulatory Commission. TIC: 247227. QAP-2-0, Rev. 5. Conduct of Activities. Las Vegas, Nevada: CRWMS M&O. ACC: MOL.19980826.0209. QAP-2-3, Rev. 10. Classification of Permanent Items. Las Vegas, Nevada: CRWMS M&O. ACC: MOL. 199903 16.0006. ANI-WIS-MD-000008 REV 00