ML20248F118

Rev 0 to Issue Resolution Status Rept Key Technical Issue: Evolution of Near-Field Environ

ML20248F118
Person / Time
Issue date:	09/30/1997
From:	NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:
Shared Package
ML20248F090	List: ML20248F087 ML20248F118
References
REF-WM-11 NUDOCS 9806040111
Download: ML20248F118 (83)

Text

. ,

ISSUE RESOLUTION STATUS REPORT l KEY TECHNICAL ISSUE: EVOLUTION OF THE NEAR-FIELD ENVIRONMENT Division of Waste Management l Office of Nuclear Material

Safety & Safeguards U.S. Nuclear Regulatory Commission l

l Revision 0 September 1997 90 60 g 971107 Wtt-11 PDR

CONTENTS Section Page ACKNOWLEDGMENTS iii 1 INTRODUCTION . . 1 2 KEY TECHNICAL ISSUE AND SUBISSUES . . .3 3 IMPORTANCE TO REPOSITORY PERFORMANCE . .5 3.1 U.S. Department of Energy Waste Containment and Isolation Strategy .5 3.2 Importance of Subissues to Total Repository System Performance .6 3.2.1 Importance to Performance of the Effects of Coupled Processes on Seep %e . . . . .. .. . .7 3.2.2 importance to Performance of the Effects of Coupled Processes on Waste Package Lifetime . .9 3.2.3 Importance to Performance of the Effects of Coupled Processes on Radionuclides Release . . . . . 10 3.2.4 Importance to Performance of the Effects of Coupled Processes on Radionuclides Transport . 12 3.3 Consideration of Coupled Near-Field Processes in Previous Performance Assessments . 13 3.3.1 U.S. Department of Energy Total System Performance Assessment 1993 . . . . 13 3.3.2 U.S. Department of Energy Total System Performance Assessment 1995 . , . . . . . . 14 3.3.3 Electric Power Research Institute Yucca Mountain Total System Performance Assessment .. . .. 15 3.3.4 U.S. Department of Energy Performance Assessment Overview Study on the Consequences of Cementitious Materials . . . .. 15 3.3.5 U.S. Nuclear Regulatory Commission Iterative Performance Assessment Phase 2. . . .... . .. .. . . ..... .16 3.4 U.S. Nuclear Regulatory Commission / Center for Nuclear Waste Regulatory Analyses Sensitiny Analyses . . . . . .... 16 3.4.1 Cement-Affected Near-Field Environment . . . . 17 3.4.2 Effects of Corrosion Products from Waste Packages on the Near-Field Environment ... .. . .. .. .. . ... .. . .. . 17

)

3.4.3 Conceptual Model of Waste Package Degradation - Brine Formation on Container Surface .. . . .......... ........ .. . . 18 3.4.4 Conceptual Model of Oxidation Rate Controlled Radionuclides Release 18 4 REVIEW METHODS AND ACCEPTANCE CRITERIA . . . . . . . . . .. . . . .20 4.1 The Effects of Coupled Processes on the Rate of Seepage into the Repository . . 20 l 4.1.1 Review Methods and Acceptance Criteria .. . . . . . 20 4.1.2 Technical Basis for Review Methods and Acceptance Criteria for Seepage .. . . ... . . .. . . . . . . . .. . 23

)

i L-_--_-_______-- .__ i

l Section Page j 4.1.2.1 Coupled Processes Affecting Flow of Water . . . 23 4.1.2.2 Effects of Engineered Materials on Seepage . .. . 26 4.1.2.3 Microbial Effects on Seepage . . . . . 27 4.2 The Effects of Coupled Processes on the Waste Package Lifetime . . .27 1 4.2.1 Review Methods and Acceptance Criteria .. .. .. .28 4.2.2 Technical Basis for Review Methods and Acceptance Criteria for Waste Package Lifetime . . . . . . . . . . 30 l 4.2.2.1 Coupled Processes Affecting Corrosion ... . . . 31 4.2.2.2 Effects of Engineered Materials on Waste Package Lifetime . .33 4.2.2.3 Microbial Effects on Waste Package LifeUme . ... . .35 4.3 The Effects of Coupled Processes on the Rate of Release of Radionuclides from Breached Plaste Packages . . . . . . ..... . 35 4.3.1 Review Methods and Acceptance Criteria . . . . . 35 4.3.2 Tecnnical Basi for Review Methods and Acceptance Criteria for Rate of Release . . . . .. . .. . . . 37 4.3.2.1 Environmental Effects on Spent Fuel and Borosilicate Glass Alteration 38 4.3.2.2 Effects of Engineered Materials on Release . . . .41 4.3.2.3 Radiolysis Effects on Radionuclides Release . . . .43 4.3.2.4 Microbial Effects on Radionuclides Release . . .43 4.4 The Effects of Coupled Processes on Radionuclides Transport Through Engirieered and Natural Barriers .. . . . .43 4.4.1 Review Methods and Acceptance Criteria . . . .44  ;

4.4.2 Technical Basis for Review Methods and Acceptance Criteria for Radionuclides Transport . . . .. .. .. . . .46 4.4.2.1 Radionuclides Transport Processes . . . . .. . .. .47 4.4.2.2 Effects of Engineered Materials on Radionuclides Transport . ... 54 4.4.2.3 Radiolysis Effects on Radionuclides Transport . . .. 56 4.4.2.4 Microbial Effects on Radionuclides Transport . .. . . . . . .. . 57.

5 STATUS OF ISLUr RESOLUTION AT THE STAFF LEVEL . . . .. .. . . . 58 5.1 U.S. Nuclear Regulatory Commission Review of U.S. Department of Energy Site Characterization Plan .. . . .. . .. .58 5.2 U.S. Nuclear Regu'atory Commission adit Review of U.S.

Department of Energy TSPA-95 . . ... . . .. . . . .. . 62 5.3 U.S. Nuclear Regulatory Commission Review of U S. Department of Energy Thermal Modeling and Testing Program . . .. .. . . . . .. .. 62 5.4 Status of issue Fasolution at the Staff Level . ... .. .. .... .. 63 6 REFERENCES .. .... . ..... ..... .. .... . . . . . . . .. . .. . 64 Appendix A: Draft Figure illustrating Elements of the NRC Staff's Total System Performarice Assessment . . . . . . . .. . . .. .. 79 ii i

i l

ACKNOWLEDGMENTS Drs. J. Apps, G. Cragnolino, P. Lichtner, W. Murphy, R. Pabalan, E. Pearcy, D. Pickett, and N.

Sridhar are gratefully acknowledged for preparing the technical core of this document. Special thanks are offered to Drs. D. Turner and W. Patrick for their excellent reviews.

iii I

1.0 INTRODUCTION

One of the primary objectives of the U.S. Nuclear Regulatory Commission's refocused .

. relicensing program is to direct all its activities towards resolving the 10 key technical issues (KTis) it considers to be most important to performance of the proposed high-level nuclear waste (HLW) repository at Yucca Mountain (YM). This approach is summarized in Chapter 1 of l - the fiscal year (FY) 1996 Annual Progress Report, NUREG/CR-6513 (U.S. Nuclear Regulatory l Commission,1996a). Other chapters address each of the 10 KTis by describing the scope of the issue and subissues, path to resolution, and progress achieved during FY 1996.

Consistent with 10 CFR Part 60 requirements and a 1992 agreement with the U.S. Department of Energy (DOE), staff-levelissue resolution can be achieved during the relicensing co,.sultation period; however, such resolution at the staff level would not preclude the issue l being raised and considered during the licensing proceedings. Issue resolution at the staff level during relicensing is chieved when the staff has no further questions or comments (i.e., open items), at a point in time, regarding how the DOE program is addressing an issue. There may be some esses where resolution at the stafr level may be limited to documenting a common understanding regarding differences in NRC and DOE points of view. Furthermore, pertinent, additional information could raise new questions or comments regarding a previously resolved issue.

An important step in the staff approach to issue resolution is to provide DOE with feedback regarding issue resolution before the viability assessment. Issue Resoluti>n Status Reports (IRSRs) are the pnmary mechanism that the staff wi!! use to provide feedback to DOE regarding progress toward resolving the subissues comprising the KTis. IRSRs include: (i) acceptance criteria and review methods for use in issue resolution and regulatory review, (ii) technical bases for the acceptance criteria and review methods, and (iii) the status of resolution, including where the staff currently has no comments or questions as well as where it does.

Additional information is also contained in the staff Annual Progress Report that summarizes the significant technical work toward resolution of all KTls during the preceding fiscal year.

Finally, open meetings and technical exchanges with DOE provide opportunities to discuss issue resolution, identify areas of agreement and disagreement, and develop plans to resolve such disagreements.

In soddion te providii,y ,adback, the IRSRs will serve as guidance for the staffs review of information in the DOE viability assessment. The staff also plans to use the IRSRs in the future to' develop the Standard Review Plan for the repository license application.

Each IRSR contains six sections, including an Introduction in Section 1. Section 2 defines the KTI, the related subissues, and the scope of the particular subissue or subissues that are

- addressed in the IRSR. Section 3 discusses the importance of the subissue to repository

- performance,' including: (i) qualitative descriptions, (ii) reference to a total system performance (TSP) flowdown diagram, (iii) results of available sensnivity analyses, and (iv) relationship to the DOE Waste Containment and Isolation Strategy (WCIS) (i.e., the DOE approach to its safety

. case). Section 4 provides the staffs review methods and acceptance criteris, which indicate the basis for resolution of the subissue and that will be used by the staff in subsequent reviews 1

w___ _ .___________._____._s_ . _ _ . _ _ _ _ _ _ _ _ _

as well. The staff technical basis for acceptance criteria will also be included to further document the rationale for their decisions. Section 5 concludes the report with the status of i resolution indicating those items resolved at the staff level or those items remaining open.

These open items will be tracked by the staff, and resolution will be documented in future IRSRs. Finally, Section 6 includes a list of pertinent references.

i 2

l

l 2.0 KEY TECHNICAL ISSUE AND SUBISSUES The near-field environment at the proposed repository at YM is defined from the perspective of i the NRC in terms of potentialimpact on the performance of the proposed geologic repository.

The near field is considered to be the portion of the site for which changes in the physical and chemical properties, resulting from the construction of the underground facility or from the heat generated by the emplaced radioactive waste, affects performance of the repository. The extent of the near field may vary substantially depending on the specific processes of concern.

l With respect to repository performance, large portions of the mountain may be affected significantly by some thermal-hydrologic-chemical (THC) coupled processes, whereas other processes may have effects only close to the engineered barrier system (EBS). Coupled processes considered for this IRSR are THC interrelations associated with the near field of the proposed repository at YM. Thermal-hydrologic (TH) couplings and thermal-mechanical couplings are addressed crimarily in other IRSRs.

l The objective of the Evolution of the Near-Field Enuronment (ENFE) KTl i: to assess all aspects of the evolution of the near-field geochemical environment that have the potentia; to impact the performance of the proposed repository. The near-field geochemistry will be perturbed from ambient conditions by variations in temperature and pressure associated with the heat production of the waste, introduction of foreign materials into the mountain, variations in fluid flow, and consequent chemical reactions. A primary system characteristic with the potential for myriad effects on performance is coupled processes. In particular, coupled THC and thermal-chemical processes will be the primary drivers on performance within the ENFE KTI. The consequences on performance from the evolution of the near-field environment are expressed as the results of coupled processes.

Five system attributes have been identified by DOE (U.S. Department of Energy,1996a) as being the most important for predicting the performance of the engineered and natural barriers of the proposed repository. These system attributes include the rate of seepage into the repository, waste package (WP) lifetime, rate of radionuclides release from breached WPs, and 1 radionuclides transport (RT) through engineered and natural barriers. These system attributes serve as one way to classify the effects of coupled processes on performance that result from the evolution of the near-field environment. (Tl ' fifth system attribute is dilution in the saturate j zone beinw the repository and is not " dressed in the ENFE KTl, See S _ctio.,3.1.)

The four subissues of the ENFE KTl have been constructed to address these system attributes.

Because consequences on performance from the evolution of the near-field environment are expressed as the results of coupled processes, the subissues of the ENFE KTl are :

e The effects of coupled processes on the rate of seepage into the repository e The effects of coupled processes on the waste package lifetime e The effects of coupled processes on the rate of release of radionuclides from breached waste packages 3

• The effects of coupled processes on radionuclides transport through engineered and natural barriers.

The scope of this report encompasses all four subissues. To adequately evaluate the impact of the evolution of the near-field environment on the performance of the repository requires addressing four aspects of coupled processes. Resolution of the subissues will require that each of the four aspects be adequately addressed. First, potential effects of coupled processes on performance must be identified. This first aspect has been completed for each subissue in .

this IRSR. l I

The second ast'ect that needs to be addressed is the natural system (minerals, groundwater {

and gaseous species; their masses and fluxes) and how it willinfluence and be influenced by coupled pcocesses. The site geochemistry offers a large buffering capacity that will moderate chemical disturbances. Controls on the ambient geochemistry would be expected ultimately to govern many properties of the near-field environment. Understanding these controls provides a basis for predictions o' near field effects. Furthermore, the site geochemistry poses initial a: d boundary conditions for modeling the induced evolution of the near field. Extensive data on ambient site mineralogy and rock chemistry are mainly based on studies conducted prior to construction of the Exploratory Studies Facility (ESF; Bish, et al.,1996). The second aspect has been addressed for each subissue, with the exception of the potentialimpact of microbial processes on the natural system. Microbial processes could modify the natural system by changing the rates of geochemical pf ocesses or by changing the physical or chemical properties of the near field.

The third aspect required for resolution of the subissues is to evaluate how engineered materials willinfluence and be influenced by coupled processes. The level of detail for this aspect in this IRSR is somewhat general because the impacts of engineered materials are strongly dependent on the design of the repository, and the design has not been finalized. In addition, the impacts of engineered materials on the near-field environment have not been fully evaluated in performance assessments (pas), have not been studied as part of the DOE thermal testing program, and have not been the focus of NRC activities this FY. The potential impact of microbial processes on the engineered materials has also not been addressed this FY.

Each of tnese first thr 8, a-pects bears on the four'.h. the adequacy of any representation of the effects of coupled processes in a PA. There has been no concerted effort to evaluate within a PA framework how the evolution of the near-field geochemical environment impacts performance. Thus, this systematic approach will provide a framework to determine the potentialimportance to performance of coupled processes.

4 t

3.0 IMPORTANCE TO REPOSITORY PERFORMANCE The consequences of coupled geochemical processes can affect several aspects of the proposed Yucca Mountain (YM) high-level waste (HLW) repository's performance. Ambient near-field geochemical conditions will be perturbed by variations in temperature and pressure associated with the heat production of the waste and introduction of foreign materials into the mountain. The changes in gas, water, and solid phase compositions and masses in the near field can affect hydrologic and mass transport characteristics, alteration of the waste package (WP) and waste form materials, and waste element speciation and solubility. The capability of the repository system to isolate waste will depend strongly on the near-field geochemistry.

Thus, the performance of the repository will also depend on the effects of coupled geochemical processes.

In a performance assessment (PA) framework, these effects of coupled processes ...a as modifiers to existing conceptual models. For instance, precipitation or dissolution of m6erals as a result of coupled thermal-hydrologic-chemical (THC) processes will affect pc.osity and permeability. These flow attributes already exist in PA codes. Thus, only the valuer of these parameters in PA modules will change as a result of the coupled processes. Likewise, the effects of coupled processes will modify the values of parameters used in WP, waste form (radionuclides release), and flow and transport modules of PA codes. The potentialimportance to performance of the evolution of the near-field environment (ENFE) key technicalissue (KTl) can be primarily attributed to the fact that the effects of coupled processes have not been previously evaluated to any extent within a PA framework.

The subissues of the ENFE KTl concern the effects of coupled processes on the rate of seepage, the WP lifetime, the rate of release of radionuclides, and on RT. Each of the subissues is directly related to a major system attribute of the U.S. Department of Energy (DOE) Waste Isolation and Containment Strategy (WCIS: U.S. Department of Energy,1996a),

and this relationship is discussed in more detailin Section 3.1. A discussion of the importance to performance of ecch subissue, and how the subissues and the effects of coupled processes are addressed within the staff PA framework is presented in Section 3.2. Consideration of the effects of coupled processes in the near-field environment in previous pas is addressed in Section 3.3. Finally, planned NRC sensitivity analyses of the e'fects of coupled processes on repository performanc . re oMlined in Sect.on 3.4.

3.1 U.S. Department of Energy Waste Containment and Isolation Strategy The original DOE strategy for waste containment and isolation at the YM site was presented in its 1988 site characterization plan (U.S. Department of Energy,1988). DOE updated that strategy as a result of additional site characterization data, advances in the engineered system design, and a changing regulatory framework (U.S. Department of Energy,1996a). The WCIS reflects recent site characterization information, new WP and repository der gns, more realistic performance predictions, and the assumption of a dose-based standard. The primary goals of the strategy are near-complete containment of radionuclides within the WPs for several thousand years and acceptably low annual doses to a member of the public living near the site (U.S. Department of Energy,1996a). The updated strategy continues to rely on engineered 5

1 -

and natural barriers to contain and isolate the waste from the public. Five system attnbutes have been identified as being the most important for predicting the performance of engineered and natural barriers, These system attributes are: (i) rate of seepage into the repository; (ii)

WP lifetirne (containment); (iii) rate of release of radionuclides from breached WPs; (iv) radionuclides transport (RT) through engineered and natural barriers; and (v) dilution in the l

saturated zone (SZ) below the repository. The first four system attributes are examined in detail as part of this IRSR and are the four subissues addressed in the ENFE KTI. The final i system attribute is not one of the ENFE KTl's subissues, but it may be influenced by the effects of coupled geochemical processes. The same coupled process (mineral precipitation due to rise of pore fluid temperatures) may affect both RT and dilution and is discussed in the section of this issue resolution status report (IRSR) on RT (Section 4.4.2.1).

A number of working hypotheses have been developed by DOE tn guide testing of the most important post-closure safety issues that support each of the attributes. The hypotheses provide a basis that DOE. .,an use to explain analyses related to TSP, and they can be used to organize, manage and explain the raticiale for DOE testing. Three of the hypumeses addressing seepage are also impacted by ::upid geochemical processes and are addressed in this IRSR. The th,ae seepage-related hypotheses that are influenced by evolution of the near-field environment are: (i) fracture flow occurs within a limited volume of the repository host rock at any given time; (ii) seepage into emplacement drifts will be limited to a small fraction of the incident percolation flux due to capillary forces; and (3) bounds can be placed on thermally-induced changes in seepage rates For the second system attribute, WP lifetime, one testable hypothesis is addressed in this IRSR. The hypothesis impacted by the evolution of the near.

field environment is that corrosion rates are very low at low relative humidity, and corrosion of the inner barrier is slow. For the third system attribute, radionuclides mobilization, one testable hypothesis is addressed in this IRSR. The hypothesis impacted by the evolution of the near-field environment is that radionuclides release from waste forms due to surface area exposed, 3 dissolution, colloid formation, and microbial activity will be low. Finally, for the fourth system attribute, RT, one testable hypothesis is also addressed in this IRSR. The DOE hypothesis impacted by the evolution of the near-field environment is that transport properties of both engineered and natural barriers will significantly reduce radionuclides concentrations due to >

depletion, diffusion and dispersion. As part of the RT discussion, the hypothesis contained in system attribute on dilution, that water percolating down through the repository horizon to the water table mixes with the flow in the aquifer, is also addressed.

3.2 Importance of Subissues to Total Repository System Performanc:

The ENFE KTl is currently considered to be an important factor in repository performance. The consequences of coupled processes may affect many aspects of repository performance. In addition, the same coupled process could affect different aspects of repository performance, l For instance, dissolution and precipitation of quartz or other minerals may occur both above l i

and below the repository horizon as a result of the changing thermal regime (U.S. Nuclear Regulatory Commission,1996a), and could, thus, impact both the seepage into the drifts and radionuclides transport away from the drifts. DOE will need to adequately demonstrate and quantify the consequences of coupled processes, resulting from evolution of the near field, that impact performance in its Total System Performance Assessment (TSPA). This will require that j 6

l

1 DOE consider the coupled processes interactions both within and among key elements of the I

natural and engineered subsystems of the repository.

NRC staff is developing a strategy for assessing the performance of the potential HLW repository at YM. As currently visualized by the staff, those elements of the strategy necessary l to demonstrate repository performance are defined as Key Elements of the Subsystem Abstraction (KESA). The KESA are illustrated in Figure A-1 in Appendix A. Acceptance criteria for abstracting each of these elements into a demonstration of compliance are under development.

As highhghted in Figure A-1, the ENFE is an important factor that needs to be considered in the abstraction of seven key elements of the engineered and natural subsystems. The seven KESA that the ENFE influences are: (i) WP corrosion (temperature, humidity and chemistry);

(ii) quantity and chemistry of water contacting waste forms; (iii) radionuclides release rates and solubility limits; (iv) fracture versus matrix flow; (v) spatial distribution of flow; (vi) retardation. ci fractures; and (vii) retardation M the SZ. Just as the effects of a single coupled process may impact more than one aspect of repository performance, both a coupled process and a h2SA may be incorporated into one or more of the ENFE subissues. The acceptance cnteria contained within this IRSR are subsidiary to and designed to complement the broader-level acceptance criteria for the @straction of the key elements. The importance to performance of the ENFE subissues and their relationship to the KESA are discussed more fully below.

3.2.1 Importance to Performance of the Effects of Coupled Processes on Seepage The effects of coupled processes on the rate of seepage into the repository is the first subissue of the ENFE KTl. The three KESA that are influenced by the evolution of the near-field environment within the scope of the seepage subissue are: (i) fracture versus matrix flow; (ii) spatial and temporal distribution of flow; and (iii) quantity and chemistry of water that affects the engineered barrier system (EBS). There are three main coupled processes that will occur in the near-field environment that have the potential to impact performance. Each of these processes needs to be considered in the evaluation of each of the KESA. The processes are:

dehydration of zeolitic horizons; coupled THC processes, resulting in mass redistribution of gaseous and liquid components, that affects the porosity and permeability of the natural system; and coupled THC processes at the in, ' face of the natural system and the engineered componen's.

The first coupled process that may impact performance of the repository is the potential dehydration of zeolitic minerals. Major geochemical changes in the near field are likely to depend primarily on the availability of water. Although unsaturated, the rocks at YM contain abundant water, commonly 10 percent of the rock volume. A large amount of zeolitic water is also available in certain horizons, primarily beneath the repository, that could be released at elevated temperatures. The potentialimportance of this process to performance has been recognized by DOE (Bish, et al.,1996). The dehydration of zeolites could impact the quantity of water that can intersect the repository horizon, since it reflects an additional amount of water beyond that associated with infiltration and deep percolation. The spatially varying distribution of the zeolitic horizons in YM, and the thermal-loading strategy will cause spatially and temporally variable dehydration of zeolites. Water released from the dehydration of zeolites 7

could impact both the spatial and temporal distribution of flow. Flow through these zeo!itic horizons will also be affected by the loss of host rock volume associated with the dehydration process, creating new fractures and widening existing fractures, thereby, leading to possible increases in fracture flow.

The second coupled process that may impact performance of the repository is coupled THC processes resulting in mass redistribution of gaseous and liquid components that affects the porosity and permeability structure of the natural system, and the chemistry of water intersecting the repository horizon. Given the temperature-dependent solubility of different minerals, it is possible that solutions (both liquid and gas phase) moving by thermally-driven convection will redistribute chemical components such as pH, chloride, oxygen, CO2 , silica, and calcium. Most extensive and rapid chemical reactions will occur where water evaporates, depositing solutes, and where distilled water containing dissolved CO2 condenses. Because water is drawn by capillarity into the finest pores of the rock, evaporation and precipitation may have the greatest effects in the rock matrix. However, gaseous transport of water vapor to cooler zones of condensation is ikely to occur dominantly in fractures, Therefore, , condensation of initially dilute, mildly-acidic water and mineral dissolution are likely to occur on fracture surfaces. The thermal effects on the natural system are both temporally and spatially variable as a result of repository design (edge effects) and the radioactive decay of the waste.

Extensive development of heat pipe effects and refluxing at elevated temperatures could cause changes in porosity, and solution composition over regulatory time frames of thousands of years. Because small changes in porosity can produce orders of magnitude changes in permeability (Lichtner and Walton,1994), the dissolution and transport of silica, followed by precipitation during evaporation, could modify the permeability distribution in the natural system surrounding Se repository horizon.

The coupled THC processes around the repository within the natural system will affect each KESA of the subissue. Thus, the spatial and temporal distribution of flow, the distribution of flow in fractures and matrix, and the quantity and chemistry of fluids intersecting the repository horizori will be impacted as a result of the evolution of the near-field geochemical environment.

The final coupled process within the subissue of seepage that may affect performance of the repository is likely to be spatially limited to very near the drifts of the repository. The reference design description of the ground control system indicates the use of both pre-cast and cast-in-place ennerete liners for emplacement drifts (TRW Environmental Safety Systerrs Inc.,1997).

Interaction of cement with the tuffaceous host rock and ambient pore fluids could nave an

! important effect on seepage. The chemistry of pore fluids in contact with hydrated cement phases is characterized by a persistent alkaline pH (> 10). Hyperalkaline cement pore water is thermodynamically incompatible with silica, a major component of the proposed YM repository host rock unit, the Topopah Spring Tuff. This unit is composed predominantly of alkali-feldspar,

. quartz, cristobalite, and tridymite (Bish, et al.,1996). Thus, migration of the high-pH cement pore water into the host rock is likely to result in strong alteration of the tuff by changing the porosity and permeability of the tuff (Lichtner, et al.,1997). The concrete drift liners will have mechanical stability that will likely depend on the spatially variable mechanical stability of the geologic medium. Thus, the stability of the drift liners and their interactions with the l surrounding geologic medium will impact both the spatial and temporal distribution of flow. The strnng interaction of the concrete with the tuff is also likely to influence the quantity of water that 8

l

I ~

I .

1 i

i can be transmitted into the drifts by both the matrix and fractures. Finally, the strong chemical

, impact of concrete on the natural fluids that intersect the repository horizon will almost certainly l

influence the chemistry of any fluids that interact with both the WP and waste form.

l 3.2.2 Importance to Performance of the Effects of Coupled Processes on Waste l Package Lifetime l

The effects of coupled processes on WP lifetime is the second subissue of the ENFE KTl The j four KESA that are influenced by the evolution of the near-field environment within the scope of j the WP containment subissue are: (i) fracture versus matrix flow; (ii) spatial and temporal

]

distribution of flow; (iii) quantity and chemistry of water that affects the EBS; and (iv) WP 1 corrosion (temperature, humidity, and chemistry). The first three KESA are the same key elements of abstraction that were addressed in the ENFE seepage subissue. These KESA are  ;

not discussed in detailin this section. as the consequences of these aspects will be addressed in the discussion of the WP corrosion KESA. The three coupled processes that .:re discussed in the seepage subissu also willimpact the WP containment subissue and WP KESA, but will also not be discussed in detailin this section. Ordy those aspects of the coupled processes that differ between the two subissups and their impact on the WP corrosion KESA will be presented here. One additional coupled process that may impact WP containment is the generation of natural and spontaneous self potentials.

In order to successfully abstract the WP corrosion process within a PA framework requires incorporating the various potential modes os corrosion and the functional dependence of the corrosion process on environmental factors. Both the modes and rates of corrosion are directly dependent on the pH, Cl, and oxygen content in the near field of the repository As a result, coupled THC process will assert a strong influence on WP corrosion. For instance, the low air i mass fraction that may be generated within the near field as a result of boiling (Wilder,1996; Lichtner,1997) will lower the oxygen concentration and will directly influence the possibility of dry oxidation of the WP.

The spatial and temporal distribution of WPs that could be affected by dripping water will be influenced by the coupled processes addressed in the seepage subissue. In addition, the chemistry of the water that contacts the WP will be the result of coupled THC processes with natural and engineered materials, including be+h the WP and potentially the concrete liner.

Depending on the ten perature of the WP and tne drifts, and the flux of water intwepting the WP, further concentraon of the fluid on the surface of the WP, due to evaporative effects, is possible. Thus, the WP KESA and the waste containment will be affected by the results of coupled geochemical processes in the near field of the repository.

One final coupled process that could affect the WP corrosion KESA is the potential corrosion of WPs by self potentials. It has been suggested that natural and spontaneous electrical potentials, known as self potentials, may generate electrical currents that may affect the performance of the waste containers (Wilder,1996). These self potentials are generated as a result of the movement of water and solutes in natural geothermal environments. Large self ]

potentials (> 500 mV) have been measured during the single heater test within the Exploratory Studies Facility (ESF) at YM. The potentialimpact of this coupled process at the repository I

,n 9 l

_____________________A

scale has not yet been addressed and no effort to determine the importance of this alternate conceptual model for corrosion of the WP has been completed.

3.2.3 Importance to Performance of the Effects of Coupled Processes on Radionuclides Release The effects of coupled processes on radionuclides release is the third subissue of the ENFE KTI.

The three KESA that are influenced by the evolution of the near-field environment within the scope of the radionuclides release subissue are: (i) WP corrosion (temperature, hu.T.'dity, and chemistry); (ii) quantity and chemistry of water contacting waste forms; and (iii) radionuclides release rates and solubility limits. The first two KESA are the same key elements of abstraction that were addressed in the ENFE seepage and containment subissues. These KESA are not discussed in detailin this section, as only the consequences of these aspects will be addressed in the discussion of the radionuclides release rates and solubility limits KESA. The three coupled processes that were discussed in the seepage subissue also will impact the radionuclides release subissue and release rate KESA, but will also not be discussed in detail in tins section.

Only those aspects of the coupled processeo tha. differ between the subissues and their impact on the radionuclides sclease rate KESA will be presented here.

Prior to water contacting the spent fuel, degradation of the cladding must occur. If credit for cladding were to be taken in PA, the effects of coupled processes on the cladding would need to be considered. Environmental variables controlled by coupled geochemical processes occurring in the near-field environment are presented next.

Above a certain critical potential, Zircaloy is susceptible to pitting corrosion in chloride-containing environments (Cragnolino and Galvele,1977). Such a potential can be naturally attained under slightly oxidizing conditions (i.e., in the presence of Fe'*). Under the environmental and potential conditions leading to pitting, stress corrosion cracking (SCC) of zirconium and Zircaloy occurs in the presence of an applied stress (Cox,1990). Whereas a decrease in Eh protects the fuel cladding from localized corrosion and SCC, it can promote failure by delayed hydride cracking. The presence of the fluoride anion in the environment, although its concentration is relatively low, may increase the uniform dissolution of zirconium alloy as a result of the greater stability of the ZrF.2. complexes compared to that of the passive ZrO, film.  ;

Corrosion of the UO2penets by contact with the groundwater, as modified by chemical and physical interactions in the near field, is the most important process affecting the long-term performance of this waste form. One of the major near-field environment factors affecting UO2  ;

waste form performance is the redox potential or Eh. The redox potential of the near-field

]

environment will be influenced by both radiolysis and the air mass fraction in the WPs.

Radiolytic oxidizing species such as OH', where "" denotes a free radical, H2 O2 . HO,', and 02' (Spinks and Woods,1976) could oxidize reduced species (e.g., Fe in the Fe container of the WP to Fe2

• and Fe5 *, N2 (aq) to NO,' or NOi, or U'* to U'*).

The pH has an effect on the rate of dissolution of spent fuel that depends on the pH range.

Under oxidizing conditions, only a slight dependence of corrosion rate on pH has been observed at pH values lower than 4, whereas at pH values between 4 and 8, the rate decreases

! 10

linearly with pH (Grambow,1989). At higher pH values, the rate of dissolution seems to be unaffected by pH changes. The nature of the anionic species present in the groundwater and their concentrations are extremely important in determining the rate of corrosion of spent fuel.

Anions such as CO 32', that form stable soluble complexes with U(6+) cations, sut;stantially increase the rate of oxidative dissolution (Blesa, et al.,1994). Corrosion is accelerated by anions in the sequence Cl~ < PO,2~ < SO,2 < F~ < CO3 2, although in the case of PO,$' and SO42 , a maximum in the corrosion rate is observed at intermediate concentrations (about 1.5 x 10 2 M) (Blesa, et al.,1994).

The modification of the pH of the leachate attributed to the formation of HNO3 by alpha-radiolysis of humid air, as well as the generation of formate and oxalate from inorganic C, may raise the solubility of actinides (Finn, et al.,1994a). Through interactions with oxidizing components, including radiolytic products, spent fuel will eventually oxidize, forming a large quantity of uranyl (UO22*)-bearing solids. Other species such as SiO2(aq), H 3SiO, , and H2 SiO,2 , can react with UNI) to precipitate complex uranyl silicates, which may tend to reduce the corrosion rates and e<posure of fresh surface by forming a protective layer over the spent fuel. Therefore, secondary oxidation products wi.i accumulate and uranyl minerals will have a larp effect on near-field physical and chemical conditions.

Secondary U phases are likely to have severalimportant effects on the near-field environment including: (i) physical disruption of structural components (e.g., cladding or degraded containers), due to the large volume increase accompanying oxidation and hydration of UO2 : (ii) plugging porosity and reducing permeability because of volume expansion; (iii) incorporation by coprecipitation or sorption of plutonium, and other radioactive waste species, that will exist as trace components in the altered system relative to U, iron, and possibly other components from the engineered and geologic system such as nickel, aluminum, and silica; (iv) limiting ingress of water and oxidants to unaltered wastes; and (v) controlling by solubility or dissolution rate, the source term for radionuclides (not just U) releases from the breached WPs. With regard to long-term performance of the proposed repository, secondary alteration products resulting from interactions of spent nuclear fuel with the near-field environment, rather than unaltered spent fuel, will control releases of many radionuclides from the WPs.

The second main waste form planned for the proposed repository at YM is borosilicate glass.

The environmental factors affecting the gene 11 or !ocalized dissolution rate of borosilicate 2

glasses are pH, fluoride, and Fe

• Ultimate ' ass waste form alteration product _e ';kely to be clay or zeolite minerals, analogous to alteration products of the natural volcanic glasses existing at YM. However, they are likely to incorporate augmented quantities of components of the EBS such as Fe and Ca. Clay minerals generally have low solubilities. Some quantity of radioactive waste species are likely to be incorporated in mineral alteration products of glass waste forms.

Interactions between cementitious materials and the near-field system can be potentially l beneficial foi mitigating release of radionuclides The persistent alkaline pH (>10) characteristic l of pore fluids in contact with hydrated cement phases favors precipitation of a wide variety of

l. radionuclides, including transuranic (Glasser, et al.,1985; Atkins, et al.,1990). On the other hand, alkaline conditions can be detrimental to the stability of nuclear waste glass. For instance, experiments by Heimann (1988) indicated that cement and glass interaction leads to 11 l

• . \

l l

accelerated dissolution and alteration of the nuclear waste glass compared to a system without cement present.

Thus, the effects of coupled geochemical processes on radienuclide releases from the repository appear likely to be of paramount importance to the performance of the proposed YM repository and will need to be carefully considered in the abstraction of radionuclides release rates.

3.2.4 Importance to Performance of the Effects of Coupled Processes on Radionuclides Transport The effects of coupled processes on RT is the last subissue of the ENFE KTI. The three KESA that are influenced by the evolution of the near-field environment within the scope of the RT subissue are: (1) fracture versus matrix flow; (2) retardation in fractures; and (3) retardation in the SZ. The three coupled processes that were discussed in the seepage issue are equally important for this subir:ue. Each of these processes needs to be considered in the evaluation of each of the KESA.

A major concern for PA in the near field is the transport of radionuclides through the EBS and the geologic setting as gaseous species, as species in colloidal form, or dissolved in aqueous solution. Each RT process is influenced by several geochemical parameters; thus, an assessment of the relative importance of each process will depend on the specific geochemical and hydrologic characteristics of the near-field environment. One mechanism for removing radionuclides from solution is the precipitation of stoichiometric radioelement compounds or coprecipitation as impunties in other minerals. Changes in system chemistry parameters, such as Eh, pH, and component concentrations, influence the solubility of radionuclides-beanng minerais.

Another retardation mechanism is sorption, and this is strongly controlled by several geochemical parameters such as solution pH. For example, sorption of actinide species, such as UO22*, NpO2 *, and Am3 *, on oxides and oxyhydroxides through a surface complexation mechanism is characterized by a sharp sorption " edge', where sorption increases sharply with increasing pH from essentially zero over a relatively narrow pH range; the location of the sorption edge differs for different actinides. The amount of radionuclides sorbed is also dependent on the rac' r. 'x!ide concentration and the number of available sorption sites. Cor 2 2 clay and zeolite minerals, sorption af radionuclides such as Cs*, Sr *, and UO2

• can also occur through an ion exchange mechanism, which is depenaent on the nature and concentration of competing cations present in solution.

Localized reducing conditions could be promoted by near field hydrologic effects and phase variations. Gas flow from the near field driven by vaporization of water is predicted to be away from the near field in all directions (Pruess, et al.,1990; Tsang and Pruess,1987), and would tend to purge air containing O2 from the near-field environment. Because the vapor pressure of water at temperatures above 95 *C exceeds the hydrostatic pressure of less than 0.1 MPa at YM, the gas phase in the near field would tend to be dominated by H2 0. Diffusion of air toward zones of relatively high water vapor pressure could reintroduce oxygen to the near field (Tsang and Pruess,1987). Local fluctuations of reducing and oxidizing conditions in the near field due 12 d

4 1

1 3

_ _ _ . - - - - " - ~ _ - - - '

. i to an unstable hydrologic regime could also induce secondary chemical effects, such as the formation of colloids (Buddemeier and Hunt.1988; McCarthy and Zachara,1989).

The oxidation state in the near field may affect sorption behavior. For example, under oxidizing conditions, technetium is principally present as pertechnetate (TcOi) and does not sorb strongly, whereas, under reducing conditions, Tc" is predominant and sorbs more strongly I (Lieser and Bauscher,1988). Elevated temperatures expected for the near-field environment may also affect sorption, but there are few data for evaluating the magnitude of the effect. In the near field, boiling in response to thermalloading would tend to partition "CO2 into the gas phan, enhancing gas transport.

The porosity and permeability could be enhanced by the dissolution of the primary minerals and reduced by precipitation of secondary minerals. Given the temperature-dependent solubility of different minerals, it is possible that thermally-convecting solutions will dissolve and redistribute minerals such as silica and calcite.

The potentialimportance of colloids in the transpo1 of radionuclides is also of con 3rn in PA in the near field. The stability of the colloidal suspension of charged particles varies as a function of ionic strength, solution chemistry, and pH. Colloid transport through the near field may be retarded in several ways, including through interaction with cementitious materials (Savage, i 1997).

Thus, the effects of coupled geochemical processes on RT will be important to the performance of the proposed YM repository and will need to be carefully considered in the abstraction of RT.

3.3 Consideration of Coupled Near-Field Processes in Previous Performance Assessments Some limited consideration of the effects of coupled processes on the performance of the proposed YM repository have been included in past PA studies. Those recent PA studies that have addressed the effects of coupled geochemical processes include: (i) TSPAs performed by DOE; the 1993 Total System Performance Assessment (TSPA-93) (Wilson, et al.,1994) and the 1995 Total System Performance Assessment (TSPA-95) (TRW Environmental Safety Systems, Inc.,1995); (ii) one TSPA prepared by the Electric Power Research Institute--Yucca Mountain Total Systerr r 3rformance Assessment, Phase 3 (Kessler and McGuire,199'i , one sensitivity stuuy by DOE, Status / Summary Reprt for Fiscal Year 1996 Activities Within the Performance Assessment Overview Study on the Consequences of Cementitious Materials (TRW Environmental Safety Systems, Inc.,1996a); and one TSPA prepared by NRC - Iterative Performance Assessment, Phase 2 (IPA Phase 2) (Wescott, et al.,1995). The manner in which these studies incorporate the effects of coupled processes is summarized in the following sections.

3.3.1 U.S. Department of Energy Total System Performance Assessment 1993 The TSPA-93 described as part of its source-term model a near field geochemistry module containing geochemistry parameters for use in the container-failure, cladding-failure, and I waste-form dissolution modules. The input paramete.s in the near field geochemistry module l

! 13 i

are pH, Eh, chloride, fluoride, and carbonate concentrations. However, in TSPA-93, Eh and chloride concentration were not used. Along with temperature from the temperature module, and fractional time the WPs are wet from the near-field hydrology module, the pH, fluoride, and carbonate concentrations were used in estimating the rate of container corrosion, cladding failure rates, and alteration rates of the waste form. Although these chemical parameters could have been allowed to vary as a function of time, in TSPA-93, these parameters were held constant for each simulation. The pH, and fluoride and carbonate concentrations used were from wells J-13 and Ue25p#1. It is recognized that these values are appropriate for far-field conditions, but it is not known how well the parameters represent conditions in the near-field environment.

Consequently, although the coding architecture was available for simulating some limited near-field conditions, only the temperature-time history can be considered a reasonable approximation for the near field.

With regard to radionuclides solubility, it v'as stated that " increased temperature L.o the repository may cause r.. ore aggressive grounowa'er chemistries and increased ^ solubilities for

. radionuclides in the t .J field; however, when the solute is transported out of the near field, the potatially lower solubilities in the far field would cause precipitation and thus would be the limiting Detor. The experts have made this assumption primarily because the dearth of information about the near-field water chemistry makes accurate predictions of solubility impossible for this region. A potential concern must be mentioned with regard to this assumption. The high thermal loads being considered for the potential repository (e.g.,114 kW/ acre) may cause near-field conditions to extend throughout the unsaturated zone (UZ:

Wilson, et al.,1994; page 9-3)."

3.3.2 U.S. Department of Energy Total System Performance Assessment 1995 Possible geochemical variations in near-field environmental conditions were considered.

However, minimal effects of changes in the geochemical environment were employed in the performance calculations. For example, with regard to solubilities, it was stated,"Because the actual changes to the near-field environment are not yet well-defined, incorporation of such effects either into [ solubility) distributions such as those discussed above or into models for predicting the solubility'-controlling phases for each radionuclides is not currently possible" (TRW Environmental Safety Syste i ; inc.,1995; page S-11). Although some pH-dependent s >lubility relations were derived, it was concluded that, "Although the derived functions inc7 orate

- pH-dependence explicitly, the near-field pH evolution is uncertain to the extent that adequate constraints do not exist for making a pH choice other than a random selection from a distribution (TRW Environmental Safety Systems, Inc.,1995; page 9-25)." For alternate solubility models, only pH 7 was considered.

In general, solubilities used in TSPA-95 were highly uncertain, which is partly represented by

' distributions spanning many orders of magnitudes (TRW Environmental Safety Systems, in::.,

1995). Comparisons to solubilities for selected elements determined in independent computations using EO3 (Wolery,1992) for ranges of possible geochemical conditions revealed that most TSPA-95 solubilities were comparable or higher (more conservative). Two exceptions are radium (Ra) and tin (Sn). Calculated solubilities of RaSO, and cassiterite 14 I

l

(SnO2 ) for a range of possible water chemistries and temperatures were near 104 m. This value is near the upper limit of the TSPA-95 range for Ra and 10 times the upper limit of the TSPA-95 range for Sn. However, considering RaSO, or SnO2 to limit solubility is perhaps unnecessarily conservative, because these trace metals are likely to be incorporated as minor components of other phases.

While not addressed in terms of effects of coupled processes, fracture and matrix interactions were analyzed as part of the PA (TRW Environmental Safety Systems, Inc.,1995). DOE incorporated the effect of matrix diffusion in their conceptual rnodel for flow and transport in the unsaturated zone. Support for that conceptual model from experimental or field data was not provided (U.S. Nuclear Regulatory Commission,1996b). Hydrological and geochemical data from the vicinity of the YM site and from analog sites suggest that matrix diffusion type processes may have limited effect on the rate of radionuclides migration (U S. Nuclear Regulatory Commission,1996b; Baca and Jarzemba,1997).

Gas phase transport or radionwlides was not evaluated in terms of performance (individual dose), since DOE judged it to be insignificant to performance. This may be a reasonable I conclusion, however, no calculations were provided to support this assertion. In addition, the rationale given in the executive summary was vague and did not reflect the incomplete rationaie presented in the body of the report.

3.3.3 Electric Power Research Institute Yucca Mountain Total System Performance Assessment The Electric Power Research Institute (EPRI) has evaluated some of the processes that the ENFE KTl has responsibility for addressing. EPRI used a code called IMARC (Integrated Multiple Assumptions and Release Calculations) to assess the performance of the individual components that contribute to the performance of the repository system (Kessler and McGuire, 1996). This study evaluated the impact of microbial processes on WP containment, and the impact of transport processes (diffusion and advection) through the concrete barrier (invert of the drift) on performance.

3.3.4 U.S. Department of Eriergy Performance Assessment Overview Study on the Conecquences of Cementitious Materials An overview study to address potential postclosure pedormance issues concerning use of large quantitbs of cementitious materials within the potential waste emplacement drifts was conducted by DOE (TRW Environmental Safety Systems, Inc.,1996a). Preliminary consequence sensitivity analyses using a TSPA model(the RIP code) were conducted. The

' impact of these engineered materials on the potential radionuclides release and transport was addressed. The analyses included evaluation of enhanced solubility of radionuclides (Np, Am, and Pu) and reduction of sorption in the UZ due to alkaline fluids that may result from the use of cementitious materials. Three cases for no retardation over distances of 10 m,100 m, and the entire UZ were evaluated. Preliminary findings from this study indicate that for migration of hyperalkaline fluids, order-of-magnitude increases in peak dose and substantial shifts to earlier times may result due to: (i) negation of sorption throughout the entire UZ; (ii) increased 15 l

-- I

I concentrations of Np and Pu; and (iii) combination of no retardation for 10 or 100 rn into the geosphere and increased concentrations of Np and Pu.

3.3.5 U.S. Nuclear Regulatory Commission iterative Performance Assessment Phase 2 An auxiliary analysis was included in Appendix K (Wesectt, et al.,1995) that addressed coupled near-field processes and their effects on the carbon geochemical system. A simple one-dimensional, uniform, time varying gas flow field was coupled to equilibrium aqueous speciation, CO2 gas evolution and transport, and calcite precipitation in a transient thermal field.

The model was used to explore possible mechanisms for "C retardation and gas phase release. However, these results were not incorporated in analyses of cumulative complementary distnbution functions.

3.4 U.S. Nuclear Regulatory Comm!ssion/ Center for Nuclear Waste Regulatory Analyses Sensitivity Aralyses The effect of coupled processes on repository performance will be assessed in terms of sensitivity to dose. This effect and the importance of parameter values assigned to physical properties in the analysis are determined by systematically performing sensitivity analyses.

Both process-level models and the abstracted models in the PA code can be used to assess the effects of coupled processes expected to be active in the near-field environment. Process-level models used by the ENFE KTl are detailed models formulated on basic principles that govern heat and mass transfer and chemical reaction for the range of expected conditions at the repository. Abstracted models within the NRC PA code [ Total Performance Assessment (TPA) Version 3.1] are designed to represent the physical processes by extracting only higher order effects identified in process-level models.

Process-level models contained within the MULTIFLO code (Lichtner and Seth,1996a,b; Lichtner,1997) have been used by the ENFE KTl to guide the input values chosen for parameters in the NRC KTis sensitivity analyses in particular, some of the input values used for the Container Life and Source Term (CLST) KTl sensitivity analyses are derived from MULTIFLO process-level modeling. Both process-level and abstracted models are used in the ENFE KTl to assess the effects of coupled processes in terms of sensitivity of dose to variations in model assumptions and parameter va'ues. j in an effort to conserve resources, some of the effects of coupled process in the near-field i environment on major components of the repository will be aodressed by the KTl responsible for the component. For instance, many of the effects of coupled processes on the WP and radionuclides release will be addressed as part of the CLST KTI's sensitivity analyses. The planned ENFE KTI's sensitivity analyses are primarily focussed on evaluating the potential dominant impact of engineered materials (cementitious materials, and steel used in WPs) may l have on the performance of the repository. Conceptual models of WP degradation and radionuclides release will also be evaluated as part of the ENFE KTl sensitivity efforts. The studies are currently underway and are described below. The results will be reported in a separate report.

16

'l

=

3.4.1 Cement-Affected Near-Field Environment Changes in water chemistry may result due to interactions between cementitious materials and l groundwater. In particular, hyperalkaline fluids (pH >10) may result from these interactions.

l These fluids are capable of precipitating radionuclides, including transuranic, thus, altering the i source term and transport behavior of radionuclides. Cement phases provide a multitude of sorption sites that could aid in retarding radionuclides migration. In addition, dissolution of the geologic barrier (e.g., tuff) by a hyperalkaline fluid could lead to a widening of fractures and enhanced groundwater flow. On the other hand, migration of a hyperalkaline fluid could result in precipitation of calcite and calcium-silica-hydrate (CSH) phases along fractures and reduce fracture porosity and hydraulic conductivity. In addition, mineral precipitation could seal fracture surfaces and enhance RT by preventing diffusion into the matrix.

Three cases simulating effects on fractures will be studied. (i) enhanced groundwater flow due to dissolution and widening of fractures; (ii) reduced groundwater flow caused by mineral precipitation in the frcctures and matrix; and (iii) reduced matrix diffusion caused by mineral precipitation along and coating fracture walls. Mo e specifically, simulat:ons will be accomplished by varying fracture permeability and porosity and matnx permeability and poroity. The potential spatial extent of the hyperalkaline plume is presently undetermined. As approximations, each case will consider one situation in which the plume is assumed to extend vertically over the lower half of the 10 m repository layer and another situatiert in which the plume is assumed to extend vertically downward from the repository horizon to the bottom of the Topopah Spring Tuff.

In addition to the above cases, formation of a hyperalkaline plume and its potential effect on RT will be studied by using radionuclides distribution coefficients, Ko s, applicable to high pH conditions. For instance, the maximum values for matrix Kos will be taken from a relevant sorption database (Bradbury and Sarott,1995).

3.4.2 Effects of Corrosion Products from Waste Packages on the Near-Field Environment I

Corrosion of waste containers could result in alteration products, such as iron oxyhydroxides. I which could either sorb and retard radionuclides migration or form pseudo-colloids that adsorb

{

and could enhance transport of . radionuclides. 3orption of many radionuclides, includir.g the actinides, on iron oxpydmxides is a strong function of pH and carbonate concentration if caroonate is present in significant quantities, such as is typically observed under ambient conditions at YM, sorption will reach a maximum at a pH that is related to the hydrolysis behavior of the radionuclides of interest. At pH values above and below this maximum, sorption will decrease to K. values close to zero.

The effectiveness of the near-field environment in physical and chemical filtering of iron oxyhydroxides from suspension will determine which role is played by steel alteration products, if the flow is predominantly through the matrix, it is likely that iron oxyhydroxides will be effectively filtered from suspension and will act as an additional barrier to transport. Fracture flow may lead to unretarded or enhanced flow, if the fracture apertures are large enough for j iron oxyhydroxide transport, and if the chemical conditions of the solution provide electrostatics 17 l l

(e.g., low ionic strength) leading to more stable suspensions. If fracture apertures are too small, or if the chemical conditions favor flocculation and settling, then iron oxyhydroxide transport will not be effective.

To investigate the potential effect of iron oxyhydroxide formation on dose, two end-member cases will be considered in TPA calculations relevant to the effects of WP corrosion products.

In one case, radionuclides will be assumed to be retarded by the iron oxyhydroxides. In the second case, radionuclides migration will be assumed to be enhanced by transport of iron oxyhydroxides. An indirect approach rnust be used in which parameters available in the TPA 3.1 code are varied to approximate the effects of these two end-member cases. As in the case for the cement-affected near-field sensitivity study. potential effects on radionuclides transport will be studied by using radionuclides distribution coefficients, Ka s, applicable to the cases. The potential spatial extent of the iron oxyhydroxide transport is not characterized. As an approximation, and to be consistent with the physical situation considered in the cement sensitivity analyses, each case will consider one situation in which the zone of transoort is assumed to extend vertically c.er the lower half of the 10 m repository layer, and another situation in which the plume is assumed te extend vertically downward from the repository horizon to the botton, of the Topopah Spring Tuff.

3.4.3 Conceptual Model of Waste Package Degradation - Brine Formation on Container Surface Aqueous corrosion is one of the chemical degradation processes that may lead to WP failure.

Water films on metal surfaces will contain a variety of components, including chloride and other soluble anionic species. Chloride ions are known to promote localized corrosion and enhance general corrosion of container materials. Evaporation of water contacting the waste container could lead to increasing chloride concentrations and eventually to brine formation on the outer overpack surface enhancing the susceptibility of both the outer and inner containers to localized corrosion.

To determine the possible effect of high chloride concentrations on dose, calculations will be done using the EBSFAll module of TPA 3.1. Waste package corrosion depends on the corrosion potential and the critical potential required to initiate localized corrosion. In EBSFAIL, the critical potential, conservatively represented by the repassivation potential, is assumed to depend on / on chlorida concentration and +ernoerature Fcr a sen temperature, localized corrosion only occurs acou a critical chloride concentration (Clb. Two cases will be considered: (i) chloride concentrations below [Ci),, for localized (pitting) corrosion; and (ii) chloride concentrations above (Clb for localized (pitting) corrosion.

3.4.4 Conceptual Model of Oxidation Rate Controlled Radionuclides Release Radionuclides release rates in the TPA 3.1 code base case scenario are based on extrapolations of the results of short-term laboratory experiments. Uncertainties in long-term extrapolations and in the dependence of release rates on environmental characteristics introduce considerable uncertainty in determination of the source term for the proposed repository at YM. An alternate approach is to use data from natural analog systems to constrain release rates on a geologic time scale. The Nopal I uranium deposit at Pena Blanca, Chihuahua, Mexico, has been studied 18 l

in detail as an analog of the proposed repository at Yucca Mountain. Using data from that site, conservative estimates of the maximum average rate of oxidation of primary uraninite (an analog of spent fuel) have been calculated (Murphy and Pearcy,1992; Murphy, et al.,1997).

This sensitivity analysis is based on the premise that the oxidation rate of spent fuelin the proposed YM repository will be comparable to the oxidation rate of uraninite at Pena Blanca.

Similarities between these sites in geology, geochemistry, climate, and hydrology support this premise. Calculation of the bounding rate of oxidation depends on the amount of uranium removed from the site. Pe6a Blanca site-specific characteristics are used to determine this quantity based on conservative estimates of the uranium solubility and the groundwater flux through the system. The maximum average uranium oxidation rate for the YM repository is determined by scaling for the masses of uranium at each site. For radionuclides incorporated in the matrix of spent fuel, the oxidation rate can be reasonably assumed to be a conservative limit on the release rate. In this sensitivity analysis, performance of the proposed repository system is evaluated, usina the alternate source term release rates based on the Pe6a Blanca natural analog studies.

l I

f 19

4.0 REVIEW METHODS AND ACCEPTANCE CRITERIA This issue resolution status report (IRSR) and its acceptance criteria focus on providing a path {

to resolution and will be revised to become more specific as the results of performance assessment (PA) sensitivity studies and other analyses and testing become available. The systematic approach to resolving the subissues is necessary because in the past there has been no concerted effort to evaluate, within a PA framework, how the evolution of the near-field geochemical environment impacts performance. As a result, the acceptance criteria are formulated in a manner that will lead to resolution of the subissues by allowing progress in determining which effects of coupled processes need to be included in any PA of the prooosed repository. It is anticipated, as the results of sensitivity studies conducted by the evolution of the near-field environment (ENFE) key technical issue (KTl) team become available, that the acceptance crteria will become more specific to each subissue. The specificity of the acceptance criteria will also reflect wh;ch processes will need to be addressed and, potentir"f' ,

the bounds of the effeus of coupled processes that need be used for an acceptable PA.

Several relatively comprehensive review documents have been published that are relevant to coupled thermal-hydrologic-chemical (THC) processes in the near field of the proposed repository at Yucca Mountain (YM) Nevada (Manteufel et al.,1993; Wilder,1996; Bish, et al.,

1996; Angell, et al.,1996). Although materialin these reports supports much of the technical basis for this IRSR, references to pnmary sources of data are generally provided in the text.

4.1 The Effects of Coupled Processes on the Rate of Seepage into the Repository U.S. Department of Energy (DOE) has identified seepage of water into waste emplacement drifts as a factor that is important to waste containment and isolation. Important factors affecting seepage are: groundwater flux at tF ipth of the repository horizon, fracture density and physical properties, presence or abse. _,f fracture coatings, rock heterogeneity, moisture content, existence of fast pathways, fluid density gradients, and others. Several aspects of seepage, particularly, the quantity and chemistry of water contacting waste forms and the spatial distribution of flow, are included as elements of the NRC PA subsystem abstraction. I Contributions will be required from several KTl groups to resolve the issue of seepage, including the Unsaturated and Saturated Flow Unaer Isothermal Conditions (USFIC) KTI, Thermal Effects on FO "(TEF) KTI, and Repositr/ Design and Thermal Mechanical Effects (RDTME) KTl, as well as the ENFE KTI. DOE must adequately estimate the quantity and chemistry of seepage and appropriately consider seepage in its assessments of waste containment and isolation. i l

4.1.1 Review Methods and Acceptance Criteria I

{

The following Acceptance Criteria apply to evaluating the DOE estimates and consideration of seepage:

• DOE included the following relevant processes and any others that may affect seepage in its analyses: (i) thermal-hydrologic (TH) effects on liquid flow; (ii) effects of fracture-matrix interaction; (iii) hydrothermal-chernical effects, such as changes in groundwater 20 e

chemistry, that may affect hydraulic properties; (iv) dehydration of hydrous phases liberating moisture and consuming thermal energy; (v) mineral and glass alteration i affecting porosity and permeability and fracture-matrix interactions; (vi) effects of l cementitious materials on hydraulic properties; and (vii) effects of microbial processes on hydraulic properties.

)

e Analyses were completed to determine effects of coupled processes on seepage.

1 e Reasonable or conservative ranges of parameters or functional relations were used to j determine effects of coupled processes on seepage. Parameter values, assumed I ranges, probability distributions, and bounding assumptions used are technically l defensible and reasonably account for known uncertainties.

e Appropriate models, tests, and analyses were used that are sensitive to the couplings under consideration. Alternative modeling aoproaches consistent with available data and current scVific understanding were investigated, and their results anu limitations were approprimly considered.

• Both temporal and spatial variations in conditions affecting seepage were considered.

e Not all couplings may be determined to be important to performance, and DOE may adopt assumptions to simplify PA analyses. However, DOE must provide a technical basis that goes beyond lack of understanding, if they neglect potenhally important couplings. Such technical basis could include independent modeling, laboratory or field data, or sensitivity studies.

e DOE's evaluation of coupled processes properly considered site charactenstics in establishing initial and boundary conditions for conceptual models and simulations of coupled processes that may affect seepage. A great amount of knowledge exists on mineralogy, petrology, and rock chemistry. Extensive water and gas chemistry data also exist in a variety of sources. Venfy that the DOE analyses of seepage are consistent with this evidence.

• DOE evaluated the effects of coupled processes that would result from repository-driven processes. Tha natural-setting data pro 'ide an indication of effects of potential coupied processes an the repository. Examples of such processes that shoulo be evaluated include: (i) zeolitization of volcanic glass, which could affect transport pathways; (ii) precipitation of calcite and opal on the footwall of fracture surfaces and the bottoms of lithophysal cavities, which indicates gravity driven flow in open fractures that could affect permeability and porosity; and (iii) potential dehydration of zeolites and vitrophyre glass, which could release water affecting heat and fluid flow.

• DOE's evaluation of coupled processes properly considered the charr.teristics of engineered materials, such as the type, quantity, and reactivity of material, in establishing initial and boundary conditions for conceptual models and simulations of coupled processes that affect seepage. Verify that the DOE analyses of the effects of coupled processes on seepage are consistent with the available information.

21

e DOE evaluated the effects of engineered materials and related coupled processes that may occur due to interactions with these engineered materials or their alteration products. Examples of such effects and processes that DOE should evaluate include:

(i) changes in water chemistry that may result from interactions between cementitious materials and groundwater, which in turn affects seepage; (ii) dissolution of the geologic barrier (e.g., tuff) by a hyperalkaline fluid that could lead to changes in the seepage properties of the geologic barrier; and (iii) precipitation of calcite or calcium-silica-hydrate (CSH) phases along fracture surfaces as a result of migration of a l hyperalkaline fluid that could affect seepage.

e DOE provided a reasonable description of the mathematical models included in their j seepage analysis and a discussion of alternative modeling approaches not considered in their final analysis.

e The mathematical models for seepage are consistent with conceptual models based l on, inferences acout the near-fhld environment, field data and natural auerations observed at the site, and expected engireered materials.

e DOE appropriately adopted accepted and well-documented procedures to construct and test the numerical models used to simulate seepage.

e Where simplifications for modeling seepage into the dnft are used for PA analyses instead of detailed process models, the bases used for modeling assumptions and approximations have been documented and justified. The model adequately incorporates important design features, physical phenomena and couplings, and uses consistent and appropriate assumptions throughout.

e Seepage models are based on the same assumptions and approximations shown to be appropriate for closely analogous natural or experimental systems. Results are verified through comparison to outputs of detailed process models and empirical observations. Model results are compared with different mathematical models to judge robustness of results.

e The initial conditions, boundary conditions, and computational domain used in  !

sensitivity analyses invntving seepage re consistent with available data.

e Mathematical model limitations and uncertainties in modeling seepage are defined and documented.

e Sensitivity and uncertainty analyses (including consideration of alternative conceptual models) should be used to determine whether additional new data are needed to better define ranges of input parameters. NRC staff will evaluate whether plans to collect these new data are adequate, and whether the data will be sufficient to support scientific conclusions about the effects of seepage on repository performance.

e ~ Data and models have been collected, developed, and documented under acceptable quality assurance (QA) procedures.

22

l .

l .

NRC staff should verify that there are no deficiency reports concerning data quality on issues related to seepage that have not been closed.

e lf used, expert elicitation were conducted and documented in accordance with the guidance in NUREG-1563 (U S. Nuclear Regulatory Commission,1996c) or other acceptable approaches.

4.1.2 Technical Bases for Review Methods and Acceptance Criteria for Seepage The technical bases for the acceptance criteria for seepage are given in this section. These bases are primarily focused on explaining why the results of different coupled processes may be important to seepage. As mentioned previously in Section 3, the dearth of analysis in past pas on the effects of coupled geochemical processes on the potential repository seepage behavior has resulted in acceptance criteria that primarily focus on ensuring that some type of analysis of the effects is completed. The sophistication of the analysis of the effect of a counted process on seepage t' at could ']e conducted and found acceptable by staff is dependent on the information available at the present, any plans to cbtain the additionalinformation as part of the long-term testing program, and the ability of PA codes to model the effects of the couplea process.

For instance, results of coupled thermal-hydrologic. geochemical modeling (see Chapter 4 of U.S. Nuclear Regulatory Commission,1996a for an example) might be used to infer the potential amounts of pnmary minerals that are either dissolved or the mass of secondary minerals that may precipitate due to the coupling. This information might be combined with existing site distribution of primary and secondary minerals (abundance, and location in the matrix or fracture or both) and the molar volume of the minerals to calculate the increase or decrease in porosity caused by the coupled THC process. If these changes in porosity are within the range of values sampled for a particular modeled hydrologic unit in the base case PA, then the effect of that coupled process on porosity need not be considered further.

j i

4.1.2.1 Coupled Processes Affecting Flow of Water 1, host rocks are silicic tuffs (70 to 80 percent SiOg (Byers,1985)). These rocks are variably vitric, devitrified to an assemblage of t 'ca minerals and alkali feldspar, or altered primarily to 'he silica-rich zeolites, clinoptilolite md mordenite, or to analcime at depth (Di;,h arid Chipera,1989). The saturated zone (SZ) groundwater is a dilute, oxidizing, sodium bicarbonate solution rich in dissolved silica. The unsaturated zone (UZ) groundwater differ substantially from SZ water, being more concentrated and dominated by calcium chloride or calcium sulfate in rocks nearer the ground surface (Yang,1992; Yang, et al., 1993,1996a, and 1996b). ,

Aqueous silica concentrations in excess of cristobalite saturation are observed in tuffaceous I aquifers at YM (Kerrisk,1987), and higher concentrations are observed in the UZ (Yang, et al.,

1996b). The high silica contents are generated by reaction of infiltrating meteoric water with Siliceous volcanic glass (White, et al.,1980). Glass alteration in the Calico Hills formation below the repository horizon is accompanied by incongruent precipitation of mordenite and clinoptitolite, zeolites with important sorptive characteristics (Ames,1964; Murphy and Pabalan, 1994). In the proposed repository horizon and the near-field environment at YM, the tuffs are l devitrified to alkali feldspar and silica mineral polymorphs (cristobalite, quartz, and tridymite).

23

Here, smectite is the dominant aluminosilicate alteration phase, occurring in abundances up to 5 percent (Bish,1988). In lower volcanic units at YM, primary glass has been altered to analcime i kaolinite rather than clinoptilolite. This mineralogic change, which is associated with the disappearance of cristobalite with depth, is consistent with a decrease in the activity of aqueous silica (Kerrisk,1983). Fracture mineralogy is characterized by calcite, smectite silica minerals, zeolites, and manganese oxides (Carlos,1989; Carlos, et al.,1991).

The gas phase in the vadose zone at YM is primarily air, approximately saturated with liquid water, and enriched in CO2 relative to the atmosphere. Gas chemistry analyses show limited variability in the CO2 content in space and time (Thorstenson, et al.,1990) The partial pressure of CO2in the gas phase has a strong effect on the pH of the coexisting groundwater,  ;

which, in turn, affects aqueous speciation, solubilities, and mineral stabilities (Arthur and Murphy,1989; Murphy,1993).

Under elevated temperatures the rates of alkali feldscar dissolution and growth of secondary phases, such as smectite, clinoptilolite, silica minerals, and calcite, would be accelerated.

Thermodynamic analyses for smectites (Ransom and Helgeson,1994) and clinopulolites (Bowers and Burns,1990) have quantified their decreasing stability with incr;asing temperature  !

and decreasing aqueous silica content. In addition, the swelling capacity of uncompacted smectite has been shown to be irreversibly decreased by alteration in a water vapor environment at temperatures above 150 =C (Couture,1985). Field evidence for temperature-induced changes can be obtained by regarding the natural environment at depth as an analog of the near-field environment (Apted,1990). Observations at YM include the transitions with increasing depth from clinoptilolite to analcime to albite and from smectite to ordered illite / smectite to illite The clay mineral data have been interpreted to give thermal profiles with temperatures ranging up to 300 'C for an extinct hydrothermal system at depth at the north end of YM (Bish,1989; Bish and Aronson,1993).

Experimental studies of hydrothermal reaction of tuffs and natural waters from YM at 90* to 250 'C (Knauss, et al., 1984,1987; Knauss,1987) show evidence for dissolution of primary minerals, precipitation of secondary phases and variations in water chemistry. Secondary precipitation of clay minerals, zeolites, cnstobalite, and calcite have been observed.

Experiments in which CO2 loss occurred showed more extensive secondary mineralization and i particularly more calcite precipitation than in pressurized closed-system experiments. Water l chemis*ry v=4tions were generally small, and achieved approximately steady-s'.e conditions )

in long-term expenments. The aqueous silica concentration was observed to increase l substantially at elevated temperatures corresponding to the increased solubility of silica  !

minerals. The water chemistry in selected experiments was reasonably represented as a function of time with partial equilibrium and kinetic reaction path models of the water-rock interactions (Delany,1985). Application of the results of these experiments to the near-field environment at YM must be judicious, because the high temperatures, high pressures, saturated conditions, and short time scales of the experiments are unrepresentative of expected conditions at YM.

! Mass transfer calculations that account for partial equilibrium and reaction kinetics in gas-water-rock interactions have provided geochemical models related to the YM site and near-field environment (Kerrisk,1983; Ogard and Kerrisk 1984; Delany,1985; Arthur and 24 l

l

Murphy,1989; Murphy,1993; Murphy and Pabalan,1994). The aqueous silica concentration and the CO2 pressure have been shown to be particularly important in defining the solid-phase assemblage and the aqueous-solution composition. At present, these models are limited

~

principally by the lack of thermodynamic and kinetic data. However, there have been several recent attempts to obtain such data by experimental and estimation techniques (e.g., Bowers and Burns,1990; Johnson, et al.,1991; Ransom and Helgeson,1994; Murphy, et al.,1996) that, in some cases, appear to yield inconsistent results Calculations of time-dependent processes are further hampered by the difficulty in realistically characterizing reactive surfaces in geologic environments. Also, coupling geochemical reactions to fluid flow is incompletely developed in models for near-field evolution.

Seepage may be affected by changes in porosity and permeability of the host rock. The porosity and permeability may be enhanced by dissolution of primary minerals that make up the matrix of the medium. Conversely, precipitation of secondary minerals may serve to plug available porosity, reducing permeability. Mino,al solubility depends on the pressure and temperature of the system of interest at well as solution pH, p(CO2 ). p(O2 ), ana salinity. Many common minerals, suco as quartz and metal sulfides, exhibit a prograde solubility, whereby precipitation is favored with decreasing temperature. Silica scale in geothermal wells is due, in large part, to the cooling of fluids that are supersaturated, with respect to silica, as they rise to the surface (Thomas and Gudmundsson,1989) In contrast, carbonates, such as calcite and

. dolomite, exhibit retrograde solubility and precipitate from solution with increasing temperature.

I Precipitation and dissolution may also be controlled by kinetic processes. For example, although calcite precipitates readily in geothernial systems. silica precipitation is kinetically controlled at temperatures below 200 'C and may not occur under supersaturated conditions (Thomas and Gudmundsson,1989).

l Gas flow from the near field, driven by vaporization of water, is predicted to be away from the near field in all directions (Pruess, et al.,1990; Tsang and Pruess,1987), and would tend to purge air containing O2 from the near-field environment. Because the vapor pressure of water .

at temperatures above 95 *C exceeds the hydrostatic pressure of less than 0.1 MPa at YM, the gas phase in the near field would tend to be dominated by H,0. Given the temperature-dependent solubihty of different minerals, it is possible that solutions (both liquid and gas phase) moving by thermally-driven convection will redistribute chemical components such as pH, chloride, oxygen, CO,, silica, and calcium.

Silica redistr:bution iri (ne (M near-field environment is likely to be controned by C,o dissolution of glass, feldspar, and cristobalite, and amorphous silica precipitation. The rate of this redistribution will depend on the aqueous silica activity and the relative rates of reaction. Silica redistribution has been observed in laboratory heater experiments with YM tuff under unsaturated conditions (Rimstidt, et al.,1989). Silica and Fe, dissolved near the heater, were transported in solution and precipitated as amorphous silica, Fe hydroxides, clay, and zeolite at the cooled end of the system. In the nonisothermal transient experiments of Lin and Daily (1990) on samples of the Topopah Spring Tuff, permeability was progressively reduced by three orders of magnitude from 1.3 x 10~" m2 to about 10" m 2due to narrowing of fracture apertures by silica deposition. Experiments of Vaughan (1987) using granite cores indicated that, although porosity was reduced by a relatively small amount, permeability was reduced by over 95 percent. Chigira and Watanab- (1994) also observed narrowing of pore throats by 25

silica precipitation in flow experiments using powdered granite and amorphous silica powder.

They calculated that at 90 C porosity would be halved in 135 yr.

Bish (1993) developed a premise that the thermal regime will induce a progressive Ostwald ripening effect in which a sequence of zeolites will form and, if enough time is available, culminate in the most stable assemblage of albite + quartz. This alteration willlead to a net volume reduction, potentially increasing porosity and permeability.

The effects of chemistry on flow are commonly neglected in TH simulations. Extensive development of heat pipe effects and refluxing at elevated temperatures could cause changes in porosity, permeability, and solution composition over regulatory time frames of thousands of years. Because small changes in porosity can effect orders of magnitude changes in permeability (Lichtner r id Walton,1994), the dissolution and transport of silica, followed by precipitation during evaporation, could modify the permeability distribution around the repository horizon. Some numerical simulations have been performed in attempting to predict the redistribution of pH, chloride, silica, and calcium in the near-field environment and tr,a effect on permeability (Lichtner and Seth,1996a; LicMner and Turner,1997).

Major geochemical changes in the near field are likely to depend primarily on the availability of water. Although unsaturated the rocks at YM contain abundant water, commonly 10 percent of the rock volume. A large amount of zeolitic water is also available in certain horizons that could be released at elevated temperatures, Most extensive and rapid chemical reactions will occur, where water evaporates, depositing solutes, and where distilled water containing dissolved CO2 condenses Because water is drawn by capillanty into the finest pores of the rock, evaporation and precipitation will occur dominantly in the rock matrix. However, gaseous transport of water vapor to cooler zones of condensation is likely to occur dominantly in fractures. Therefore, condensation of initially dilute acidic water and mineral dissolution are likely to occur on fracture surfaces. Together, these processes could lead to an increase of fracture permeability and a decrease of matrix permeability. However, if water precipitated on fracture surfaces dissolves minerals there and precipitates secondary phases with larger volumes before the water is imbibed, then fracture permeability could decrease as well. The locus of dissolution and precipitation reactions, with respect to fractures and matrix, could affect the hydrologic behavior of the near-field system, including seepage and is presently poorly constrained in coupled hydro-chemical modeling. Results from the Larr~ Bbck Test (LBT) and Drift Scale Test (DST) thermal experiments may be usefulin helping to constrain the extent of these effects.

4.1.2.2 Effects of Engineered Materials on Seapage The usa of cementitious materials, in the form of concrete inverts and linings, is being considered for the estimated 179 km of emplacement drifts, roadways for construction, and emplacement ramps and service mains. Although cement is used primarily for its structural (e.g., high compressive strength) and physical (e.g., low permeability) properties, its effect on the near-field chemical environment of a repository could be pronounced. The staff analysis of the effects of engineered materials on seepage assumes the reference case design for the proposed repository. The base case design assumes pre-cast concrete liners for drift support.

The effects of engineered materials other than cementitious material may not be impprtant to seepage, but any effects of these materials on seepage will be addressed in Revision 1 of this 26 I

l I l

IRSR. In addition, the following preliminary discussion of cementitious materials reflects only a limited effort by the ENFE KTl to address this topic during this fiscal year (FY). A more complete discussion of the effect of concrete on seepage into the drifts will be presented in Revision 1 of this IRSR.

Interaction of cement with the tuffaceous host rock and ambient groundwater could have an important effect on seepage. The chemistry of pore fluids in contact w!th hydrated cement phases is characterized by persistent alkaline pH (>10). Hyperalkaline cement pore water is thermodynamically incompatible with silica, a major component of the proposed YM repository host rock unit. Thus, migration of the high-pH cement pore water into the host rock is likely to result in strong alteration of the tuff. Preliminary calculations by Lichtner, et al. (1997) suggest that strong alteration of the tuff host rock at YM and cement in contact with the tuff could result from interaction of cement and tuff pore waters and the respective minerals. Because of the low silica concentration of the cement pore water, the host rock would begin to dissolve on contact with the hyperalkaline fluid (Lichtner and Eikenberg,1995). Dissolution of tuff could lead to widening of the fractures and enhancement of seepage and groundwater flow through the repository. As the host rock dissolved and the silica concentration increased. CSH phases would precipitate and clog the pore spaces. Alternatively, precipitation of calcite and CSn phases along the interface of the fracture and matrix could seal the fractures from the matrix, producing isolated channels through which groundwater could flow. However, if sufficient amounts of calcite and CSH phases are precipitated along fracture walls, reduction in fracture porosity and permeability, or fracture plugging, could result in diminished seepage and groundwater flow through the repository. Results of Lichtner, et al. (1997) show that porosity reduction within the tuff matrix could isolate it from fracture pore water and could affect seepage. Precipitation of calcite would also occur, as the low COrhigh Ca cement pore fluid mixes with the ambient groundwater containing high CO2 concentrations (Lichtner and Eikenberg,1995; Steefel and Lichtner,1994). Both the LBT and DST thermal tests could provide important insights that could constrain the potential effects on seepage resulting from cementitious materials interacting with the tuff.

4.1.2.3 Microbial Effects on Seepage This information will be provided in Revision 1 of this IRSR, currently scheduled for FY 1998.

4.2 The Effects of Coupled Processes 1 the Waste Package Lifetime DOE identified waste package (WP) lifetime (containment) as a key factor that is important to its waste isolation strategy. Important factors effecting containment are: (i) relative humidity (RH); (ii) seepage and the presence or absence of liquid water on the WP; (iii) WP degradation mechanisms, such as general corrosion, pitting corrosion and microbial attack; (iv) protection of the inner barrier provided by the outer barrier of a double-walled container; (v) cathodic protection of the inner barrier; (vi) WP temperature; (vii) thermal heat load of a waste container; (viii) radiolysis effects on corrosion (e.g., production of nitric acid); (ix) deposition of salts on the waste container; and (x) performance of backfill; and others. WP containment is embodied in the NRC PA subsystem abstraction element of WP corrosion. Contributions from the Container Life and Source Term (CLST) KTI, Structural Deformation and Seismicity KTI, RDTME KTI, and TEF KTl will be required, in addition to those of the ENFE KTl to resolve this subissue. DOE

.7

must adequately estimate the WP lifetime and appropriately consider processes affecting containment in their assessments of waste containment and isolation.

4.2.1 Review Methods and Acceptance Criteria The following Acceptance Criteria apply to evaluating the DOE estimates and consideration of WP lifetime:

e DOE included the following relevant processes and any others that may affect WP lifetime in their analyses: (i) TH effects on WP lifetime; (ii) hydrothermal-chemical effects, such as changes in groundwater chemistry that may affect WP lifetime; (iii) dehydration of hydrous phases liberating moisture that may affect WP lifetime; (iv) effects of cementitious materials on WP lifetime; and (v) effects of microbial processes on WP lifetime.

e Analyses were completed to determine effects of coupled processes on WP ufetime.

e Reasonable or conservative ranges of parameters or functional relations were used to determine effects of coupled processes on WP lifetime. Parameter values, assumed ranges, probability distributions, and bounding assumptions used are technically defensible and reasonably account for known uncertainties.

e Appropriate models tests, and analyses used are sensitive to the couplings under consideration. Alternative modeling approaches consistent with available data and current scientific understanding are investigated. and their results and limitations are appropriately considered.

e Both tempora' and spatial variations in near-field conditions on WP lifetime are considered.

e Not all couplings may be determined to be important to performance, and DOE may adopt assumptions to simplify PA analyses. However, DOE must provide a technical basis that goes beyond lack of understanding, if they neglect potentially important couplings. Such a technical basis could include independent modeling, laboratory or

'ield data, or ceacitivity studies.

e DOE's evaluatio,1 of coupled processes properly considered site characteristics in establishing initial and boundary conditions for conceptual models and simulati; 1s of coupled processes that may affect WP lifetime. A great amount of knowledge exists on mineraltgy, petrology, and rock chemistry. Extensive water and gas chemistry data also exist in a variety of sources. Verify that DOE's analyses of WP lifetime are consistent with this evidence and couplings that result.

e DOE evaluated the effects of coupled processes that have been well-documented at YM and that would result from repository-driven processes. The natural setting and thermal test data provide an indication of effects of potential coupled processes on the repository. Examples of such processes that should be evaluated include: (i) 28 I

5

zeolitization of volcanic glass, which cou'd affect water chemistry and WP corrosion processes; and (ii) generation of self potentials in the near-field that could affect WP degradation.

• DOE's evaluation of coupled processes properly considered the characteristics of engineered materials, such as the type, quantity, and reactivity of material, in establishing initial and boundary conditions for conceptual models and simulations of coupled processes that affect WP lifetime. Verify that DOE's analyses of the effects of coupled processes on WP lifetime are cons: stent with the available information.

• DOE evaluated the effects of engineered matenals and related coupled processes that may occur due to interactions with these engineered materials or their alteration products. Examples of such effects and processes that DOE should evaluate include changes in water chemistry that may result from interactions between cementitious materials and groundwater, which, in turn, affects WP lifetime.

e DOE provided a reasonable description cf the mathematical models included in their WP lifetime analysis and a discussion of alternative modeling approaches not considered in their final analysis.

• The mathematical models for WP lifetime are consistent with conceptual models based on, inferences about the near-field environment, field data and natural alterations observed at the site, and expected engineered materials.

• DOE appropriately adopted accepted and well-documented procedures to construct and test the numerical models used to simulate WP lifetime.

• Where simplifications instead of detailed process models are used for PA analyses, the bases used for modeling assumptions and approximations have been documented and justified. The model adequatelf incorporates i important design features, physical phenomena, and couplings, and uses consistent and appropriate assumptions throughout.  !

• Models are based on the same assumptions and approximations shown to be appropriate for closelv analogous natural or experimental systems. Results a.a verified through compJascri to outputs of detailed process models and empirical observations.

Model results were compared with diffeient mathematical models to judge robustness of results.

• The initial conditions, boundary conditions, and computational domain used in the l sensitivity analyses are consistent with available data.

• Mathematical model limitations and uncertainties in modeling WP lifetime are defined i and documented. l

• Sensitivity and uncertainty analyses (including consideration of alternative conceptual I models) should be used to determine whether additional new data are needed to better l

29 l

define ranges of input parameters. Staff wm evaluate whether plans to collect these new data are adequate, and whether the data will be sufficient to support scientific conclusions about the effects of WP lifetime on repository performance.

• Data ar,d models have been collected, developed, and documented under acceptable QA procedures.

• NRC staff should verify that there are no deficiency reports concerning data quality on issues related to WP lifetime that have not been closed.

• If used, expert elicitation were conducted and documented in accordance with the guidance in NUREG-1563 (U.S. Nuclear Regulatory Commission,1996c) or other acceptable approaches.

4.2.2 Technical Bases for Review Methods and Acceptance Criteria for Waste Packaae Lifetime The technical bases o the acceptance criteria for WP lifetime are given in this section. These bases are primarily focused on explaining why the results of different coupled processes may be important to WP lifetime. As mentioned previously in Section 3, the dearth of analysis in past pas on the effects of coupled geochemical processes on the potential repository WP lifetime has resulted in acceptance criteria that primarily focus on ensuring that some type of analysis of the effects is completed. The sophistication of the analysis of the effect of a coupled process on WP lifetime that could be conducted and found acceptable by staff is dependent on the information available at the present, any plans to obtain the additional information as part of the long-term testing program, and the ability PA codes to model the effects of the coupled process.

For instance, it has been suggested that natural and spontaneous electrical potentials, known as self potentials, may generate electrical currents that may affect the performance of the waste containers (Wilder,1996). These self potentials are generated as a result of the movement of water and solutes in natural geothermal environments. Large self potentials (> 500 mV) have been measured during the single heater test within the Exploratory Studies Facility (ESF) at YM. The relationship between the self potentials that are generated as a result of differences in the diffuse double layer at the rock-water interface and the corrosion potential at the conte 5er surface, which is a mixed potential due to electron exchange reactions, is not clear at present. )

Detailed ana.yses are needed to establish the importance of the self-potentials for the corrosion of containers, It would be necessary for DOE to either collect data at a larger scale, such as in the drift scale heater test, to indicate that this potential process is unimportant on the scale of l the repository, or DOE would be required to present a detailed technical calculation, supported I by existing data, that assessed whether the measured self potentials could lead to corrosion of container material. If its analysis should indicate that self potentials may be important to WP l performance calculations, then it would be expected that information would be collected to I support the abstraction of the process, and the analysis of the effects of the process, within a i

PA framework.

30 1 ,

4.2.2.1 Coupled Processes Affecting Corrosion The ability to calculate moisture redistribution at the dnft scale is essential to determine how fast the WP vrill corrode. At this scale. the geometry of the individual WP becomes important, unlike the repository-scale modelin which the waste is assumed to be distributed uniformly.

There have been several attempts to model the drift scale (Nitao,1988; Pruess, et al.,1990).

The problem is difficult because of the variation in scale within the computation domain.

Symmetric boundary conditions are usually imposed, implying an infinite array of evenly spaced WPs.

Container performance under dry conditions is dictated by the WP temperature. Two factors that determine container life under these conditions, in particular, that of the steel overpacks, are the rate of oxidation in hot, dry air and thermal embnttlement due to long-term aging at temperatures well above 100 C. The rate of uniform oxidation for pure, polycrystalline Fe in air has been empirically determined to have inverse logarithmic kinetics below 200 *C (Fchiner, 1986) In addition to miform oxidation, intergranular oxidation may be possible at low temperatures due to greater diffusivity of oxygen siong grain boundaries. Howeves, both these processes, to a first approximation, are dependent mainly on temperature and impuntics in the {

steel, and not on the chemical composition and physicochemical properties of the near-field environment, with the exception of the oxygen partial pressure.

l l

Elevated temperatures in the near field at YM are expected to lead to important geochemical changes. Localized reducing conditions could be promoted by near-field hydrologic effects and phase vanations. Gas flow from the near field, dnven by vaporization of water, is predicted to be away from the near field in all directions (Pruess, et al.,1990; Tsang and Pruess,1987), and would tend to purge air containing O2 and CO2 from the near-field environment. Because the vapor pressure of water at temperatures above 95 'C exceeds the hydrostatic pressure of less than 0.1 MPa at YM, the gas phase in the near field would tend to be dominated by H2 0.

Estimates of the temporal extent of this period of reduced air mass fraction, based on thermal-hydrologic modeling, range from hundreds of years to a few thousand years, depending on the thermalloading of the repository (Wilder,1996; Lichtner,1997). The validity of the model prediction of reduced air mass fraction could be tested as part of the DST thermal test in the ESF, and the results could he p to constrain the extent of the affect on the near field of the repository.

A variety of metal alloys that are thermodynamically unstable in contact with oxidizing water are being considered as container materials for the YM repository. Although corrosion of these materials may be slow, it would consume oxidants in the near-field environment. The oxidation of Fe in the repository near field can create a local decrease in Eh. Provided the air mass fraction in the near field remains high, the extent of the reduced zone may not be large for various reasons: (i) oxygen reduction on Fe is irreversible (far from equilibrium) and is diffusion limited; (ii) corrosion of Fe in an oxidizing environment leads to the formation of Fe oxides and oxyhydroxides that can further decrease the rate of electrochemical reduction of O2 , and (iii) the initial formation of a-FeOOH can lead to a secondary reduction reaction to Fe30 , which is  !

argued to be the case for alternating wet and dry environments (Nishikata, et al.,1994). During the dry period, Fe

• oxides or oxyhydroxides are oxidized by air to a-FeOOH, and the cyclic process proceeds due to the electronic conductivity of the inner layer of Fa30. 4 Therefore, 31

1 .

corrosion of container materials may lead to locally-reducing conditions in the near field and strong gradients in oxidation potential, despite the prevailing oxidizing nature of the geologic setting.

The time period at which the RH exceeds a critical value, RH c, is an important factor in determining the container performance. The corrosion rates in humid air at RH>RH c, as well as in liquid, aqueous environments, are higher than those in dry air. The time at which rewetting of the containers occurs depends on the near-field environment. In addition to higher corrosion rates, greater complexity is exhibited by humid air and aqueous corrosion processes depending on the near-field chemistry due to the formation of different corrosion products.

The outer steel overpack of the WP is included in the design as a corrosion allowance material that is expected to undergo slow, uniform corrosion. However, under some circumstances, the steel can undergo localized corrosion or environmentally assisted cracking:

• Localized corrosion in the form cf pitting or crevice corrosion may occur, wnen the external environment is moderately a!Aaline.

e Localized corrosion may occur, when there is alternating wetting and drying of a region of the WP, for example, by dripping from a fracture.

• Stress corrosion cracking (SCC) can occur at a pH of about 10, and when the corrosion potential (related to the Eh of the environment) reaches a entical value.

The geochemical parameters that affect container performance via localized corrosion, SCC, and hydrogen embrittlement modes of degradation include pH, Eh, temperature, CI', NOi, and HCOi. Uniform corrosion is not included in this discussion because it is not considered as important in determining container performance as localized corrosion or SCC. It must be noted that. in mildly alkaline environments (pH ranging from 8 to 11), carbon steel can undergo localized rather than uniform corrosion because a protective passive film is formed on the metal surface. This passive film slows down uniform corrosion rates by several orders of magnitude, but makes the metal more prone to the localized breakdown of passivity. Should the pH of the water contacting containers be neutral (pH =7) or acidic then uniform corrosion will be important to WP lifetime calculations.

The various environmental factors affect the tabure processes differently. For exmople, an increase in the redox potential of the environment, En, can increase the susceptibility of a material to localized corrosion and SCC (Cragnolino, et al.,1996) but can decrease the susceptibility to hydrogen embrittlement. The relationships for the various failure modes and the effect of temperature are valid for aqueous conditions within the range of 25 to 100 C (Angell, et al.,1996). The relationships indicated by (Angell, et al.,1996) are necessarily simplified and examples may be found that violate the relationships indicated within specific combinations of potential and pH. One example, the domains for SCC of carbon steelin the presence of different anions, has been presented and discussed by Angell, et al. (1996)in terms of a potential-pH diagram adapted from Ford (1983) with certain modifications (Sridhar, et al.,1994).

32

r ,

Another potential failure mode is thermally-induced fractures that may occur if a high thermal loading strategy is adopted. Fracture of the steel overpack can occur under a combination of mechanical stresses and thermal embrittlement. This failure process does not lead to a modification of the environment. However, it will promote an earlier contact of the geochemical  !

I

' environment with the inner overpack. This is an example of a process that may require temporal considerations in the analysis of the geochemical effects on containment.

Under circumstances of localized corrosion, the environment experienced by the inner overpack

]

and other WP materials is the environment inside the localized corrosion areas or cracks on steel. The evolution of this environment is affected by a combination of: (i) corrosion of steel ,

that releases Fe2* ons; (ii) reduction of oxygen that diffuses from the outside; (iii) reduction of water or H* inside the crevice created by the crack or localized corrosion front; (iv) the hydrolysis of Fe2 *; and (v) the electromigration of anionic species, such as Cl and SO,2 , into the crevice and transport of various ionic species out of the localized corrosion or cracked region. It has been predicted that, depending upon the rates of Fe oxidation to Fe 2* and H 2O or H* reJuction, the pH inside the ;revice aan be acidic or alkaline (Sridhar, et al., sw6).

Changes in the crevice environment will also depend on the external environment, the presence of air, moisture film, and oxide scale on the steel surface. The increase in acidity ana Cl concentration inside fissures has been observed in the case of meteoritic irons (Buchwald and Clarke,1989). Increased Cl concentration, up to one weight percent of corrosion products, has been reported on land-based Fe archeological objects, and up to 13 weight percerit Cl has been found in marine-based artifacts (Turgoose, 1982,1989). The increase in Cl concentration in akaganeite (G-FeOOH) formed in crevices is attnbuted to the disintegration of meteoritic objects over long time periods (Organ,1977).

Penetration of the Alloy 825 (or other Ni-base alloys) inner overpack will lead to further acidification as a result of the hydrolysis of Cr'* formed by localized dissolution. Experimental evidence for acidification within crevices of Alloy 825 has been documented (Sridhar and Dunn, 1994). At higher temperatures, crevice pH approaching one may be found in cracks or pits in Alloy 825. It has been shown that the presence of Mo, which is added to increase the corrosion resistance of these alloys, can decrease the pH further, depending upon the potentialinside the crevice.

Thus, depe1 ding unon the rate of movement , nd dripping of water into the container, the water in contact with the containers may become ac,oic, leading to enhanced solubility M actinides.

However, galvanic coupling of the inner container with the outer overpack may decrease the corrosion potential established at the Alloy 825 surface below the repassivation potential for localized corrosion, precluding the occurrence of this phenomenon and, therefore, the generation of acidic conditions. Nevertheless, galvanic coupling requires a good electronic conductivity and a low-resistance electrolytic path to be effective. In this regard, water content and its ionic strength are critical factors.

4.2.2.2 Effects of Engineered Materials on Waste Package Lifetime Staff analysis of the effects of engineered materials on containment assumes the reference case derign for the proposed repository. The base case design assumes pre-cast concrete 33 1

~. ,

liners, rather than carbon steel ribbing, for drift support. The effects of engineered materials other that cementitious material are likely to be important to containment, however only a preliminary discussion is presented below. In addition, the following preliminary discussion of cementitious materials reflects only a limited effort by the ENFE KTl to address this topic during this FY. Thus, a more complete discussion of the effect of engineered materials on containment will be presented in Revision 1 of this IRSR.

Predicting the geochemical effects of the introduction of engineered materials in the near-field environment spans the engineering and geoscience disciplines. Stability in the repository can be enhanced by the use of engineering materials that are stable in some natural environments, such as copper in a low-sulfur, reducing environment, or by introduction of materials in the near field that react geochemically to improve isolation (Langmuir,1987). A variety of metal alloys that are thermodynamically unstable in contact with oxidizing water are being considered as container materials for the YM repository. In addition, if carbon steelis used as structural

, support for the drifts, then reactions affecting the containers would also affect the structural supports. Although corrosion of these materials may be slow, it would consume oxidants in the near-field environment and thus, could affect the continued corrosion of the WP. Therefwe, corrosion of containe ..aterials may lead to locally reducing conditions in the near field and strong gradients in oxidatio,1 potential, despite the prevailing oxidizing nature of the geologic setting.

It is important to predict changes in near-field chemistry as affected by cementitious materials over the time period of regulatory :nterest, particularly with respect to solution pH, which is a key parameter controlling container corrosion. Alkaline conditions provide an environmen: that results in the formation of a tightly-adhering, passive film, thought to be y-Fe2 O3 , on carbon steel, a material that may be used as an overpack on waste canisters, and, thereby, passivates the steel and protects it from uniform corrosion. On the other hand, localized corrosion of carbon steel may occur in the form of pitting or crevice corrosion when the external environment is moderately alkaline (pH-8 to 10). SCC can occur in a HCO3 /CO3 '

environment at a pH of about 10, when the corrosion potential (related to the Eh of the environment) reaches a cntical value. However, the presence of cementitious materials will tend to keep HCOi/CO 2 3 concentrations low.

A study by Atkinson, et al. (1989) indicated that interaction of groundwater typical of a clay environment with cement couki maintain a pH atove 10.5 for a time period on the order of a ,

few hundred thouso, d years, under the low fic'.. rates assumed in that study. hsvever, results of these types of studies are highly dependent on the assumptions used in the calculations, such as groundwater flow rates, amount of cementitious materials present in the repository, and stability of the CSH gel. Simple extrapolation of results from experiments using laboratory-aged cement pastes is likely to be invalid because the solid and aqueous chemistry of cements will change considerably within the relevant timeframe (10 to 10,000 yr), even in a closed system (Atkins, et al.,1991). For example, the same study by Atkinson, et al. (1989) indicated that, if recrystallization of the CSH get occurred in the long term, lower pH could result due to the lower solubility of the (.,rystalline CSH phases. Formation of crystalline CSH phases by recrystallization of pre-existing CSH gel is likely in a high-level waste (HLW) repository, due to the long timeframe involved and tne elevated temperatures imposed by radioactive decay heat from emplaced nuclear wastes. Even a modest temperature excursion to 55 *C for 6 to 12 34  !

_ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - _ _ _ _ _ _ _ _ _ _ - - - - i

months can result in partial transformation of CSH gel to mc. etable, though poorly crystallized, phases, such as jennite and tobermonte (Atkins, et al.,1994). Thus, modeling of cement interactions with the near-field environment and its potential effect on WP lifetimes must consider the likelihood that cement chemistry is dominated by phases other than those present in the initial material. Although a number of simulations of the evolution of cement-pore fluid i

and some simulations of groundwater-cement interactions have been conducted using estimated data (Glasser, et al.,1987; Atkinson, et al.,1989; Reardon,1992; Lichtner and Eikenberg,1995; Neall,1996), most of these simulations were conducted for 25 *C and assumed the presence of amorphous CSH gel. Thus, the results may not be relevant to cement-water interactions in e HLW repository It is important to note that ongoing laboratory experiments by DOE and results from both LBT and DST thermal tests should help to constrain the potentialimportance of cementitious materials on the performance of the WP.

4.2.2.3 Microbial Effects on Waste Package Lifetime This information will be provided in Revision 1 of this IRSR, currently scheduled for FY 1998.

4.3 The Effects of Coupled Processt.s on the Rate of Release of Radionuclides fmm Breached Waste Packages DOE has identified radionuclides mobilization from the waste form as a key factor affecting dose.

Radionuclides release rates and solubility limits constitute one of the NRC key elements of subsystem abstraction for PA. Contnbutions from the CLST KTl will be required in addition to those of the ENFE KTl to resolve this subissue. DOE must adequately estimate rates of radionuclides release and appropriately consider these rates and processes affecting them in its assessments of waste containment and isolation.

4.3.1 Review Methods and Acceptance Criteria The following Acceptance Criteria apply to evaluating the DOE estimates and consideration of rate of release:

e DOE is to include the following processes and any others that may affect rate of release in its analyses: (i) TH effects on liquid flow; (ii) hydrothermal-chemical effects such as changes in groundwater chemisy that may affect rate of re! ease; (iii) dehydratim r tydrous phases libaratir., moisture and consumin; then % erergy; (iv) mineral and glass alteration affecting rate of release; (v) effects of cementitious materials on chemical conditions and hydraulic properties affecting rate of release; and (vi) effects of microbial processes on rate of release.

e Analyses were completed to determine effects of coupled processes on rate of release.

Reasonable or conservative ranges of parameters or functional relations were used to determine effects of coupled processes on rate of release. Parameter values, assumed ranges, probability distnbutions, and bounding assumptions used are technically defensible and reasonably account for known uncertainties.

35

O

• Appropriate models, tests, and analyses were used that are sensitive to the couplings under cor, sideration. Alternative modeling approaches consistent with available data and current scientific understanding are investigated, and their results and limitations are appropriately considered.

• Both temporal and spatial variations in conditions were considered.

e Not all couplings may be determined to be important to performance, and DOE may adopt assumptions to simplify PA analyses. However, DOE must provide a technical basis that goes beyond lack of understanding if they neglect potentially important couplings. Such a technical basis could include independent modeling, laboratory or field data, or sensitivity studies.

• DOE's evaluation of coupled processes properly considered site characteristics in establishing initial and boundary conditions for conceptual models and sim...ations of coupled processes that may affect rate of release. A great amount of knowledoe exists on mineralogy, petrology, and rock chemistry. Extensive water and gas-chemistry data also exist in a variety of sources. DOE's analyses of . ate of release should be consistent with this evidence.

• DOE has evaluated the effects of coupled processes that have been well-documented at YM, and that would result from repository-driven processes. The natural setting data provide an indication of effects of potential coupled processes such as zeolitization of volcanic glass, that could affect water chemistry, in turn, affecting radionuclides release.

• DOE's evaluation of coupled processes properly considered the characteristics of engineered materials, such as the type, quantity, and reactivity of material, in establishing initial and boundary conditions for conceptual models and simulations of coupled processes that affect rate of release. DOE's analyses of the effects of coupled processes on rate of release should be consistent with the available information.

e DOE has evaluated the effects of engineered materials and related coupled processes that may occur due to interactions with these engineered materials or their alteration oroducts. Examples of such effects and processes that DOE should evaluate include changes in ad.er chemistry that may .$sult from interactions between cementtaus materials and groundwater, which, in tu.n, affects rate of release.

e DOE has provided a reasonable description of the mathematical models included in their rate-of-release analysis and a discussion of alternative modeling approaches not considered in its final analysis.

e The mathematical models for rate of release are consistent with conceptual models based on, inferences about the near-field environment, field data and natural alterations observed at the site, and expected engineered materials.

• DOE appropriately adopted accepted and well-documented procedures to construct and test the numerical models used to simulate rate of release.

36 i

e Where simplifications instead of detailed process models are used for PA, the bases used for modeling assumptions and approximations have been documented and justified. The model adequately incorporates important design features, physical phenomena and couplings, and uses consistent and appropriate assumptions throughout.

e Models are based on the same assumptions and approximations shown to be appropriate for closely analogous natural or experimental systems. Results are verified through comparison to outputs of detailed process models and empirical observations.

Models results have been compared with different mathematical models to judge robustness of results.

e The initial conditions, boundary conditions, and computational domain used in the sensitivity analyses are consistent with available data.

e Mathematical model limitations and uncertainties in modeling the rate of release are defined and documented.

• Sensitivity and uncertainty analyses (including consideration of alternative conceptual models) should be used to determine whether additional new data are needed to better define ranges of input parameters. Staff will evaluate whether plans to collect these new data are adequate, and whether the data will be sufficient to support scientific conclusions about the effects of rate of release on repository perfonnance e Data and models have been collected, developed, and documented under acceptable QA procedures e NRC staff should verify that there are no deficiency reports concerning data quality related to effects of coupled processes on the rate of radionuclides release that have not been closed.

e if used, expert elicitation were conducted and documented in accordance with the guidance in NUREG-1563 (U.S. Nuclear Regulatory Commission,1996c) or other acceptable approaches.

4.3.2 Technical tsases for Review Methods and Acceptance Criteria for kate of Release The technical bases for the acceptance criteria for rate of radionuclides release are given in this section. These bases are primarily focused on explaining why the results of different coupled processes may be important to the rate of radionuclides release from the near-field environment.

As mentioned previously in Section 3, the dearth of analysis in past pas on the effects of coupled geochemical processes on radionuclides release has resulted in acceptance criteria that primarily focus on ensuring that some type of analysis of the effects is completed. The sophistication of the analysis of the effect of a coupled processes on the rate of release that could be conducted and found acceptable by staff is dependent on the information available at the present, any plans to obtain the additional information as part of the long term testing program, and the ability of PA codes to model the effects of the coupled process.

37

_ _ _ _ _ _ _ _ _ _ _ . _ _ _ 1

1 For instance, high pH solutions may be generated due to interaction of water with the concrete and this may significantly alter the solubility of radionuclides. DOE's assessment of this impact would be acceptable if it demonstrated using geochemical equilibrium modeling codes, such as  !

E03 (Wolery,1992) or others, that the solubility range chosen for its reference case PA was conservative relative to the results of the equilibrium calculations. ,

4.3.2.1 Environmental Effects on Spent Fuel and Borosilicate Glass Alteration Near-field environmental factors, including Eh, pH, temperature, Cl, NOi, and HCOi, and F' affect degradation modes of spent fuel and its cladding. Two of these degradation modes are related to processes that affect the integrity of the Zircaloy fuel cladding. Zirconium alloys are susceptible to a form of hydrogen embrittlement called delayed hydride crack..'g. This phenomenon i: promoted by the precipitation of brittle zirconium hydndes (ZrH2.-) in areas of stress concentrations upon cooling from high temperature (Cox,1990). Slow cooling may induce reorientation of plate-like hydrides into an axial rather than circumferential distributi. .

facilitating failure (Chan,1996). Although cladding creep at moderate temperatures is not dependent on environmental factors, it is car, sider ed a plausible mode of failure (Santanan, et al.,1992). Above a wrtain critical potential, Zircaloy is susceptible to pitting corrosion in chloride-containing environments (Cragnolino and Galvele,1977). Such a potential can be naturally attained under slightly oxidizing conditions (i.e., in the presence of Fe *) Under the environmental and potential conditions leading to pitting, SCC of zirconium and Zircaloy occurs in the presence of an applied stress (Cox,1990). Whereas a decrease in Eh protects the fuel cladding from localized corrosion and SCC, it can promote failure by delayed hydride cracking.

Despite its relatively low concentration, the presence of the fluoride anion in the environment may increase the uniform dissolution of zirconium alloy as a result of the greater stability of the ZrF.2 complexes compared to that of the passive ZrO2 film Localized reducing conditions could be promoted by near field hydrologic effects and phase variations. Gas flow from the near field, driven by vaporization of water, is predicted to be away from the near field in all directions (Pruess, et al.,1990; Tsang and Pruess,1987), and would tend to purge air containing O2 from the near-field environment. Because the vapor pressure of water at temperatures above 95 C exceeds the hydrostatic pressure of less than 0.1 MPa at YM, the gas phase in the near field would tend to be dominated by H2 0.

Corrosion of the UO,9lets by contact with the 3roundwater, as modified by chemical and physical interactions in the near field, is the most important process affecting the long-term performance of this waste form. One of the major influential factors determined by the near-field environment is the redox potential or Eh. Eh generally increases by gamma- or alpha-radiolysis. However, the most significant but related factor determining the corrosion rate of spent fuelis the corrosion potential E . Since UO .,2 is a relatively good electronic conductor because of its deviation from stoichiometry, E,is a well-defined electrochemical parameter for spent fuelimmersed in an aqueous environment. The rates of reduction of species such as 0 2 and H2O2 are coupled to the rate of oxidation of UO,.,, establishing E, as a mixed potential on the interface between the oxide and solution (Shoesmith, et al.,1989). The effect of the potentialis important due to the oxidative nature of the dissolution of UO 2.

38 l

l

The pH has an effect on the rate of dissolution of spent fuel that depends on the pH range.

l Under oxidizing conditions, only a slight dependence of corrosion rate on pH has been observed at pH values lower than 4, whereas at pH values between 4 and 8, the rate decreases linearly with pH (Grambow,1989). At higher pH values, the rate of dissolution seems to bc unaffected by pH changes. As in the case of other metals, valuable information can be compiled in terms of Eh-pH diagrams for the U-H2 O system in the presence of ceriain anions l (Paquette and Lemire,1981) over which specific domains for the dominant degradation modes can be superimposed. Temperature increases the rate of dissolution of UO 3. although the functional dependence is not well established over a wide range of temperatures.

The nature of the anionic species present in the groundwater and their concentrations are extremely important in determining the rate of corrosion of spent fuel. Anions such as CO,2 ,

that form stable soluble complexes with U(6+) cations, substantially increase the rate of oxidative dissolution (Blesa, et al.,1994). At low CO 32 concentrations (0.001 M), the rate of dissolution is proportional to the otal concentration because the rate-determining s'an is thc surface complexation of CO3 2 (Blesa, et al.,1994L At intermediate concentration (0.5 M), the rate depends on the square root of the total concentration because solution transport of CO 32 to the surface is rate controlling, or dissolution of an initially-formed UO2 CO3 film controls the overall rate (Grambow,1989). At a higher CO 32 concentration (1.0 M at 100 C), the corrosion rate reaches a constant value, but at even higher concentrations, the rate decreases, probably due to the formation of surface films (Needes, et al.,1975). These concentrations, while high for the norninal water composition, may occur due to evaporative processes. Corrosion is accelerated by anions in the sequence Cl < PO 43 < SO,2 e F < CO 2-3 . although in the case of PO,' and SO 2, 4 a maximum in the corrosion rate is observed at intermediate concentrations (about 1.5 x 10 2 M) (Blesa, et al.,1994).

Other species, such as SiO2(aq), H 3SiO, , and H 2SiO 2 4 , can react with U(VI) to precipitate complex uranyl silicates, which may tend to reduce the corrosion rates ano exposure of fresh surface by forming a protective layer over the spent fuel. Under certain circumstances, acceleration of spent fuel dissolution can occur as a result of spallation of the alteration layers.

Rapid increases in the concentration of spent-fuel dissolution products may lead to the saturation of the medium with secondary alteration products, accelerating the precipitation of secondary phases and, eventually, the piefere,,tial release of certain radionuclides. Bates, et al. (1995) fwod that irv emittent at.ditions or untrolled amounts of groundwater to spet fuel led to precipitation of most of the transuranic elements (Am, Cm, and Pu) with the exception of Np. The fractional release follows the sequence Np=Cm>Am>Pu without exhibiting an increase with time, whereas Cs release was much greater and increased with time. Wilson (1990) used semi-static experiments (i.e., involving periodic removal of a leachant aliquot and replacement with fresh solution) and observed that actinide (U, Pu, Am, Cm, and Np) concentrations reached constant values rapidly, suggesting that steady-state conditions between spent-fuel dissolution and secondary-phase formation are established. Formation of U(6+) secondary phases, such as uranophane, was confirmed. Lower actinide concentrations, with the exception of Np, were measured at 85 *C than at 25 *C, suggesting that the solubility limiting phases are formed more rapidly at the higher temperature. Alternatively, this effect could be the consequence of the retrograde solubility of secondary products. The presence of Pu, Am, 39

i i l

l and Cm as colloids in the leachates was reported, but the formation of precipitated secondary phases predominated at 85 C.

Under oxidizing conditions and in the presence of carbonate anions, there is a large driving force for the dissolution of the UO2 matrix. Soluble radionuclides, such as "7Cs, "Sr, and '2sSb, exhibited congruent dissolution from spent 'uel in flow-through tests, and the release rate of ,

these fission products decreased with time to a steady-state value similar to the release rate of l U from the UO, matrix (Gray, et al.,1992). In semi-static tests, the fractional release of "Sr, )

'37Cs,1291, and "Tc increased with temperature and almost linearly with time. Species such as I 2

Ca . y 2*g , H3 SiO, , H2 SiO,2, and CO 32 precipitated from solution in tests conducted at higher l

temperatures. Bates, et al. (1995) suggested that the corrosion rate of the matrix and release

{

of radionuclides are accelerated in unsaturated tests compared to those under semi static

{

conditions, leading to incongruent release of individual fission products and actinides, probably I controlled by the formation of particulate in solution. In addition, the corrosion rate, and j especially the rate of rad *iclide .elease, is very dependent on the characteris* cf the spent )

fuel (i.e.. composition, degree of burnup). Reoxidation of the fuel was not considered to be a

{

factor in the acceleration of the dissolution rate (Bates, et al.,1995), but the modification of the pH of the leachate attributed to the formation of HNO3 by alpha-radiolysls of !'umid air, as well j as the generation of formate and oxalate from inorganic C, may raise the solubility of actinides (Finn, et al.,1994a). All these effects are postulated to become even more important at {

j relatively large surface area-to-groundwater volume ratios such as may be expected in the UZ l at YM However, acid generating processes may be counteracted by alkalinity deriving from cemeat-water interactions.

Through interactions with oxidizing components, including radiolytic products, spent fuet will eventually oxidize forming a large quantity of uranyl (UO22*)-bearing solids. Natural analog (Pearcy, et al. 1994) and experimental (Wronkiewicz, et al.,1992) studies indicate that schoepite. soddyite, and uranophane are among the secondary minerals likely to form from spent-fuel oxidation. Furthermore, these studies indicate that rates of oxidation of reduced uraninite and unirradiated fuel (both analogs of spent fuel) are rapid relative to transport of U away from the natural geologic setting or the experimentally simulated WP, respectively.

Therefore, secondary oxidation products will accumulate and uranyl minerals will tave a large effect on near-field physical and chemical conditions.

Secondary U phases are liKei- to have severa, ,moorta ,t effects on the near-fi ;a environment including: (i) physical disruption of structural components (e.g., cladding or degraded containers), due to the large volume increase accompanying oxidation and hydration of UO2 ; (ii) plugging porosity and reducing permeability because of volume expansion; (iii) incorporation by coprecipitation or sorption of plutonium, and other radioactive waste species, that will exist as trace components in the altered system relative to U, iron, and possibly other components from the engineered and geologic system, such as nickel, aluminum, and silica; (iv) limiting ingress of water and oxidants to unaltered wastes; and (v) controlling by solubility or dissolution rate the source term for radionuclides (not just U) releases from the engineered barrier system (EBS).

With regard to long-term performance of the proposed repository, secondary alteration products resulting from interactions of spent nuclear fuel with the near-field environment, rather than unaltered spent fuel, will control releases of many radionuclides from the EBS.

l 40 e

1 s Experimental (Holland and Brush.1980) and thecret cal (Murphy,1997) studies indicate that the solubilities of likely uranyl minerals, such as schoepite and uranophane, are retrograde with temperature. The repository honzon will reach its maximum temperature shortly after waste l emplacement (e g., within tens or hundreds of years), and thereafter it will experience an l- environment of continuously decreasing temperature. Consequently, the solubilities of alteration products of spent fuel willincrease with time. In contrast, through a process of Ostwald ripening, increasingly stable secondary phases, with lower solubilities, will crystallize.

The second main waste form planned for the proposed repository at YM is borosilicate glass.

The environmental factors affecting the general or localized dissolution rate of borosilicate glasses include Eh, pH, temperature, Cl, HCOi, H3 SiO6 F , and Fe2

• As in the case of metals, the interrelationship of Eh and pH on the dissolution of waste glass can be displayed in a potential-versus-pH diagram (Jantzen,1992). In general, Eh has practically no effect on the dissolution of the glass matrix because silicon, boron, and aluminum, which are the principal network formers of borosilicate glasses, do not undergo changes in oxidation state within the range of expected Eh "alues ader repository conditions. The effect of pH is far more l

important. The rate of dissolution is strongly accelerated at alkaline pH due to matrix dissolution. At pH lower than 4, the rate is accelerated by diffusion-controlled alkali ion exchange for hydrogen ions. Whereas many anions have a minor effect on the solubility and rate of dissolution of borosilicate glasses, fluonde accelerates the dissolution substantially through the formation of SiF.2 complexes. The value of RH is important in the durability of g! asses in humid air. Glasses are also susceptible to environmentally-assisted cracking in aqueous environments (McCauley,1995), but the effect of this phenomenon on radionuclides i i

releases may be far less important than that associated with generalized dissolution.

Alteration of glass depends pnmanly on the activity of aqueous silica. In the ambient geochemical environment and for predicted geochemical conditions in the host rock, the aqueous silica concentration is large (Yang,1992; Murphy,1993) A glass waste form would be expected to be fairly unreactive for these conditions, although other components of the glass ]

(e g., B) will also affect its stability. Alkalinity produced by interactions of water with cementitious materials, and lowered silica activity, as a consequence of precipitation of silicates by interactions of groundwater and unstabie engineered materials, could enhance the alteration of glass waste forms. Ultimate glass waste form alteration products are likely to be clay or zeolite minerals, analogous to alteration products of the natural volcanic glasses existing at YM.

However, . hey are Ukely to incorporate augmented quantities of components of the EB0 coch as Fe and Ca. Clay minerals generally have low solubilities. Some quantity of radioactive waste species are likely to be incorporated in mineral alteration products of glass waste forms.

4.3.2.2 Effects of Engineered Materials on Release Staff analysis of the effects of engineered materials on radionuclides release assumes the reference case design for the proposed repository. The base case design assumes pre-cast concrete liners, rather than carbon steel ribbing, for drift support. The effects of engineered materials other that cementitious material are likely to be important to radionu lide release, however only a preliminary discussion is presented below. In addition, the following preliminary discussion of cementitious materials reflects only a limited effort by the ENFE KTl to address j 41 l

(

. I this topic during this FY. Thus, a more complete discussion of the effect of engineered materials on radionuclides release will be presented in Revision 1 of this IRSR.

A variety of metal alloys that are thermodynamically unstable in contact with oxidizing water are being considered as container materials for the YM repository. In addition, if carbon steel used as structural support for the drifts, then reactions affecting the containers would alto affect the structural supports. Although corrosion of these materials may be slow, it would consume oxidants in the near-field environment and thus could affect the continued corrosion of the spent fuel. Therefore, corrosion of container materials may lead to locally-reducing conditions in the near field and strong gradients in oxidation potential, despite the prevailing oxidizing nature of the geologic setting.

Corrosion products from metallic components, mostly in the form of metal cations, can affect corrosion rates directly through precipitation reactions forming secondary minerals that may slow the rate of dissolution. Conversely, corrosion rates can be increased by indirect action of corrosion products that may change the redox potential and the pH of the environment The redox potential can increase by the action c.,f redu;ible cations such as Fe'*, whert.as the pH can decrease by the nydrolysis of highly-charged cations, such as Cr 3* and Fe 2+, among others.

In addition, the presence of low molecular weight organic compounds including carboxylic acids produced by degradation of fuel, lubricants, or other organic materials, either by chemical or biochemical mediated processes. may accelerate the rate of corrosion of spent fuel due to the formation of complexing species, especially those able to form chelates.

The use of cementitious materials, in the form cf concrete inverts and linings, is being considered for the estimated 179 km of emplacement drifts of the proposed YM HLW repository. The use of these materials is in addition to the planned use of cement in roadways for construction, and emplacement ramps and service mains and is discussed in the Advanced Conceptual Design report (TRW Environmen:al Safety Systems, Inc.,1996b). Although cement is used primarily for its structural (e.g., high compressive strength) and physical (e.g., low permeability) properties, its effect on the near-field chemical environment of a repository could be pronounced. Cements are extremely fine-grained, high-surface area materials containing somewhat soluble and thermodynamically metastable phases (e.g., a gel-like chase designated CSH because it contains Ca, Si, and H2 O) that are unstable with respect to crystalline cement phases. These properties and the partially interconnected pore network of the solids make these rnterials potentially reaMive with the near-lield environment and the EBS Interactions between cementitious materials and the near-field system can be potentially beneficial for mitigating release of radionuclides. The persistent alkaline pH (>10) characteristic of pore fluids in contact with hydrated cement phases favor precipitation of a wide variety of radionuclides, including transuranic (Glasser, et al.,1985; Atkins, et al.,1990). For example, interaction of cement with aqueous U6* can result in the formation of Ca-bearing phases uranophane or becquerelite, a poorly-crystallized Ca-uranyl hydrate (Atkins, et al., 1988,1990).

On the other hand, alkaline conditions can be detrimental to the stability of nuclear waste glass.

For example, experiments by Heimann (1988) indicated that cement-glass interaction leads to accelerated dissolution or alteration of the nuclear waste glass compared to a system without cement present.

42 l

4.3.2.3 Radiolysis Effects on Radionuclides Release Aside from autoradiolytic effects on pr > release waste characteristics, consideration of radiolysis is most important in its potentially complex effects on aqueous oxidation-reduction conditions.

Although radiation may affect prefailure chemistry outside the WP, depending on package shielding, to an extent that corrosion rates are affected (Reed and Van Konynenburg,1993),

this section is concerned with post-failure effects on radionuclides mobilization because of the massive multi-purpose container design.

According to Dubessy, et al. (1988), the dose of absorbed y rays is only 0.02 times the dose of absorbed a particles in a given time. Also, according to Spinks and Woods (1ir76) (cited in Dubessy, et al.,1988), a single 1 MeV a particle can ionize 10 5molecules as it loses energy.

Therefore, the primary cause of water radiolysis is a-particle radiation. Radiolysis occurs close I to the site of radioactive decay and can affect wetted surfaces of radioactive waste forms.

Radiolytic oxidizing species, such as OH', where "" denotes a free radical, H2 C 2 , IO 2', and O2' (Spinks and Woods.1C76), could oxidize reduced ipecies (e.g., Fe in the WP to Fe 2+ and Fe'*,

N2 (aq) to NOi or NOi, and U" to U5*). Molecular hydrogen (H 2) produced as the result d the combination of two H', in contrast, is relatively non-reactive, and is likely to diffuse away from the site of radiolysis. Various experimental studies using gamma radiation suggest that radiolysis will promote waste form (both spent fuel and glass) instability arid radionuclides mobility through both enhancement of oxidative processes and lowering of pH (Wronkiewicz, et alc,1991,1993; Sunder, et al.,1992; Sunder and Christensen,1993). As pointed out by Van Konynenburg (1986), such processes are enhanced by unsaturated conditions expected in the proposed YM repository. On the other hand, bicarbonate could depress the radiolytic pH lowering (Van Konynenburg.1986), as could cement-water interactions.

4.3.2.4 Microbial Effects on Radionuclides Release This information will be provided in Revision 1 of this IRSR, currently scheduled for FY 1998.

4.4 The Effects of Coupled Processes on Radionuclides Transport Through Engineered and Natural Barriers DOE considers radiormclide transport (RT) a hv performance attribute of the proposed repository. Retardatiw vi, radionuclides in frac:ures in tne UZ and in the SZ confute two of the NRC key elements of subsystem abstraction for PA. Contributions from the USFIC KTl and the Radionuclides Transport (RT) KTI, should it be revived in the next FY, as well as from the ENFE KTl will be required to resolve this subissue. Because many of the same geochemical concerns are common to the ENFE and RT KTis, and the RT KTl is currently inactive, a more thorough discussion of technical basis for the ENFE review methods and acceptance criteria concerned with RT is presented. It is expected that the discussion of the effects of coupled processes on RT will be revised during FY 1998 in light of activity in the RT KTI. DOE must j adequately estimate the RT characteristics of the near field and appropriately consider RT in its  ;

assessments of waste containment and isolation.

l

.o  ;

t

l 4.4.1 Review Methods and Acceptance Criteria The following Acceptance Criteria apply to evaluating the DOE estimates and consideration of RT:

• DOE included the following relevant processes and any others that may affect RT in its l

! analyses: (i) TH effects on liquid flow, (ii) hydrothermal-chemical effects, such as changes in groundwater chemistry that may affect RT; (iii) dehydration of hydrous

, phases liberating moisture that could affect flow and RT; (iv) mineral and glass l alteration, affecting porosity and permeability and flow in fractures and the matrix; (v) effects of cementitious materials on hydraulic and sorptive properties; (vi) effects of corrosion products of container materials and waste forms on RT: and (vii) effects of microbial processes on RT.

• Analyses were completed to determine effects of coupled processes on RT e Reasonable or conservative ranges of parameters or functional relations were used to determine effects of coupled processes on RT. Parameter valves, assumed ranges, probability distributions, and bounding assumptions used are technically defensible and reasonably account for known uncertainties.

• Appropriate models, tests, and analyses were used that are sensitive to the couplings under consideration. Alternative modeling approaches consistent with available data and current scientific understandings are investigated, and their results and limitations are appropriately corsidered.

e Both temporal and spatial variations in conditions were considered.

I e Not all couplings may be determined to be important to performance, and DOE may {

adopt assurnptions to simplify PA analyses. However, DOE must provide a technical basis that goes beyond lack of understanding if they neglect potentially important l couplings. Such a technical basis could include independent modeling, laboratory cr field data, or sensitivity studies.

• DCE's evalua' c of coupled procer ses properly considered site characteristics, h establishing initial and boundary condienns for conceptual models and simulations of ,

coupled processes that may affect RT. I e DOE evaluated the effects of coupled processes that have been well-documented at YM, and that would result from repository-driven processes. The natural setting data provide an indication of effects of potential coupled processes on the repository.

Examples of processes that should be evaluated include: (i) zeolitization of volcanic glass, that could affect transport pathways; (ii) precipitation of calcite and opal on the footwall of fracture surfaces and the bottoms of lithophysal cavities that indicates l gravity-driven flow in open fractures, and isolation of transport pathways from sorption sites in the rock matrix; and (iii) precipitation and dissolution of manganese oxides on 44 e

fracture surfaces, illitization of smectite, and recrystallization of zeolites to analcime, that could affect sorption characteristics.

e DOE's evaluation of coupled processes took proper consideration of the characteristics of engineered materials, such as the type, quantity, and reactivity of material, in establishing initial and boundary conditions for conceptual models and simulations of coupled processes that affect RT. DOE's analyses of the effects of coupled processes on RT should be consistent with the available information. l

• DOE evaluated the effects of engineered materials and related coupled processes that may occur due to interactions with these engineered materials or their alteration products. Examples of such effects and processes that DOE should evaluate include:

(i) changes in water chemistry that may result from interactions between cementitious materials and groundwater, which, in turn, affect RT; (ii) dissolution of the geologic barrier (e g., tuff) hv a hyperalkaline fluid that could lead to changes in the RT properties of the geologic barner; and (iii) precipitation of calcite or CSH phases along fracture surfaces as a result of migration of a hyperalkaline fluid that could affect RT.

• DOE provided a reasonable description of the mathematical models included in its RT analysis and a discussion of alternative modeling approaches not considered in its final analysis.

e The mathematical models for RT are consistent with conceptual models based on, inferences about the near-field environment, field data and naturai alterations observed at the site, and exoected engineered materials.

• DOE has appropriately adopted accepted and well-documented procedures to construct and test the numerical models used to simulate RT.

e Where simplifications instead of detailed process models are used for PA, the bases used for modeling assumptions and approximations have been documented and justified. The model adequately incorporates important design features, physical phenomena, and couplings, and uses consistent and appropriate assumptions throughout. For instance, raising the temperature of the SZ by only 15 C, could cause saturation of tuffaceous waters with respect to calcite (Murphy,1995). Precipitation of Jalcite nec '.he groundwater table could affect transport pathways or re.muat'on of radionuclides. For the example provided, thermal hydrologic modeling might be used to demonstrate that water in the SZ would not be heated by 15 C. Alternatively, if SZ temperatures were predicted to rise by 15 C, the field data (fluid chemistry and solid phase distribution) might be used to indicate where precipitation could occur.

Sensitivity analyses would then need to be completed to assess how changing the flow assumptions (porosity and permeability of both fracture and matrix, and the depth to which mixing might change as a result of calcite precipitation) affects overall system performance.

• Models are based on the same assumptions and approximations shown to be appropriate for closely analogous natural or experimental systems. Results are verified 45

~

o through comparison to outputs of detailed r ocess models and empirical observations.

DOE has compared model results with different mathematical models to judge robustness of results. For instance, in the calcite precipitation example provided in the ,

previous acceptance criterion, site characteristics constrain current groundwater flow l paths and potentiallocations for precipitation of calcite due to temperature rise of 15 I C. For the calcite precipitation example, staff would venfy that DOE evaluated a change in fracture permeability, if calcite saturation was predicted from a thermal-  ;

hydrologic analysis, and would verify that sensitivity studies had been completed that evaluated this change in terms of PA.

• The initial conditions, boundary conditions, and computational domain used in the

]

sensitivity analyses are consistent with available data.

e Mathematical model limitations and uncertainties in modeling RT are defined and documented.

e Sensitivity and uncertainty analyses (inclLding consideration of alternative conceptual models) & . d be used to determine whether additional new data are needed to better define ranges of input parameters. NRC staff will evaluate whether plans to collect these new data are adequate and whether the data will be sufficient to support scientific conclusions about tne effects of RT on repository performance.

e Data and models have been collected, developed, and documented under acceptable I

QA procedures.

e NRC staff should venfy that there are no deficiency reports conceming data quality in relation to RT that have not been closed.

e if used, expert elicitation were conducted and documented in accordance with the guidance in NUREG-1563 (U.S. Nuclear Regulatory Commission,1996c) or other acceptable approaches.

4.4.2 Technical Bases for Review Methods and Acceptance Criteria for Radionuclides Transport The technical bases - $9 review methods and acceptance criteria for RT are given in this section. These bases are primarily focused on explaining why the results of different coupled

. processes may be important to RT. As mentioned previously in Section 3, the dearth of analysis in past pas on the effects of coupled geochemical processes on the potential RT behavior has resulted in acceptance criteria that primarily focus on ensuring that some type of I

analysis of the effects is completed. The sophistication of the analysis of the effect of a coupled process on RT that could be conducted and found acceptable by the staff is dependent on the I information available at the present, any plans to obtain the additional information as part of the long-term testing program, and the ability of PA codes to model the effects of the coupled process.

46 I

i

O For instance, as a result of the water reacting with concrete as it leaves the drift, fluids along flowpaths beneath the repository horizon might have a high pH. This condition could substantially affect the values assumed for retardation in the UZ. One acceptable approach to evaluate this effect would be to conduct sensitivity studies using the DOE PA code. Sensitivity studies using the DOE PA code could be completed using values of K,s associated with cementitious repositories if substantial negative changes in the performance of the repository resulted from the use of these alternative Ka s, then it would be expected that these alternative values be used to assess performance of the repository.

4.4.2.1 Radionuclides Transport Processes A number of processes may operate in the near field to control the migration of radionuclides from the waste forms through the engineered barriers and into the geologic setting. The following is a brief summary of those processes that may be significant within the near-field environment.

A major concern for PA in the near field is the transport of radionuclides through the EBS and the geologic setting as gaseous species, or as species in colloidal form, or dissolved ir. equeous solution. Each RT process is influenced by several geochemical parameters; thus, an assessment of the relative importance of each process will depend on the specific geochemical and hydrologic charactenstics of the near-field environment However, it is useful to quaktatively desenbe the effects of changing geochemical parameters on near-field radionuclides retardation and transport processes.

One mechanism for removing radionuclides from solution is the precip'tation of stoichiometric radioelement compounds or coprecipitation as an impunty in other minerals. Changes in system chemistry parameters, such as Eh, pH, and component concentration, influence the solubility of radionuclides-bearing minerals. For example, reduction of UO/* to U" greatly reduces U in solution through precipitation of reduced U minerals such as uraninite (e.g.,

Langmuir,1987). Under oxidizing conditions, increases in dissolved silica and other species can stabilize minerals such as soddyite and uranophane. These minerals will sequester not only U, but also other actinides through coprecipitation (Murphy and Prikryl,1996).

Another retardation mechanism is sorption, vi ich it strongly controlled by several geochemical parameters such as solution pu For ev.amph sorptior, of actinides, such as UO/*, NpO,*, and Am2*, on oxides and o: y.,ydroxides through a surface complexation mechanism is characterized by a sharp sorption " edge", where ;orption increases sharply with increasing pH from essentially zero over a relatively narrow pH range; the location of the sorption edge differs 2

for different actinides. The presence of complexing ligands, such as CO 3

, decreases actinide sorption to near zero with further increases in pH (>7-9) (e.g., Kohler, et al.,1992; Pabalan and Turner,1997). The amount of radionuclides sorbed is also dependent on the radionuclides concentration and the number of available sorption sites. For clay and zeolite minerals, sorption of radionuclides, such as Cs*, Sr2 *, and UO/*, can also occur through an ion exchange mechanism, which is dependent on the nature and concentration of competing cations present in soli; tion.

47

E.

• l i l

l The oxidation state in the near field may affect sorption behavior. For example, under oxidizing l

l conditions, technetium is principally present as pertechnetate (TcOi) and does not sorb strongly, whereas, under reducing conditions, Tc" is predominant and sorbs more strongly (Lieser and Bauscher,1988). Organic substances can also influence sorption processes. For example, studies of Kohler, et al. (1992) indicate that the presence of organic ligands in millimolar concentrations (10

• to 10 M) can reduce neptunium sorption through complexing.

Elevated temperatures expected for the near-field environment may also affect sorption, but there are few data for evaluating the magnitude of the effect. Limited batch data for temperatures up to 85 C suggest that sorption coefficients for Am, Ba, Ce, Cs, Eu, Pu, Sr, and U on crushed tuff materials either remain constant or increase with increasing temperature (Meijer,1990).

In the near field, boiling in response to thermalloading would tend to partition "CO2 into the gas phase, enhancing gas transport Results of Codell and Murphy (1992) indicate that, after an early initial release of "C to the gae phase, CO2 will dissolve into the aqueous phase and calcite precipitation will serve to sequester "C w longer times. The amount of gas transport is also sensitive to the heat load imposed by the r3pesitory. The calculated releases were also dependent on the travel time to the surface, which was dependent on the Darcy velocity and the partitioning coefficient between the gas and aqueous phases. Current DOE total system performance assessment (TSPA) models do not explicitly include "CO gas transport (TRW Environmental Safety System, Inc ,1995). Other potential gas-phase species, such as *l and 8Cl, are assumed to be transported as gases without any retardation through the EBS and will then be dissolved in the aqueous phase (TRW Environmental Safety Systems, Inc.,1995).

The porosity and permeability could be enhanced by the dissolution of the primary minerals and reduced 'uy precipitation of secondary minerals. Given the temperature-dependent solubihty of different minerals, it is possible that thermally-convecting solutions will dissolve and ' redistribute minerals such as silica and calcite. This could not only affect RT in the UZ, but also in the SZ beneath the repository in addition to its effect on RT, the change in porosity and permeability of the uppermost portion of the SZ could impact the extent of any vertical mixing of fluids leaving the UZ. Simulations by Travis and Nuttall (1987) suggest that reduced permeability due to quartz reprecipitation may enhance waste isolation. In contrast, Verma and Pruess (1988) determined that silica redistribution in a hydrologically-saturated fractured medium did not have a significant effect on near-field temperatures, pore pressures, or fluid flow.

The potentialimportance of colloids in the transport of radionuclides is also of concern it. PA in the near field. The stability of the colloidal suspension of charged particles varies as a function of ionic strength, solution chemistry, and pH. Colloid transport through the near field may be retarded in several ways. If the particles are small enough to enter the porous medium, transport may eventually lead to a constriction that is too small for them to pass through. This leads to a straining of the colloids from solution. Surface filtration involves building a barrier at the interface between the water and pore when the particles are too large to enter the pores of the medium. Particles may also be removed from suspension by chemical and physical interactions with the geologic medium and the EBS. Most of the research on colloid formation and transport is based on studies of saturated media. Colloid transport at YM is likely to be complicated by the unsaturated conditions, partial coverage of pores and fracture walls, and the presence of two-phase flow. For example, a recent study on the effect of hydrologically-48

E -

l l

unsaturated conditions on colloid transport indicated that hydrophilic colloids are preferentially sorbed at the gas-water interface while hydrophobic colloids sorb at both the gas-water and solid-water interfaces (Wan and Wilson,1994).

Near-field environmental parameters including Eh, pH, temperature, Cl~, CO/', SiOw,and dissolved organic carbon, affect near-field RT. For example, a temperature increase might increase the dissolution of sorbing phases such as zeolites and oxyhydroxides, reducing sorption and degrading the performance of the EBS and the geologic setting in the near field with regard to transport. In addition, complex processes such as precipitation and dissolution may affect transport in more than one way.

Sorption The principal concern of RT in the near field is the advective transport of radionuclides dissolved in aqueous solution through the EBS to the geologic setting. Minerals in different components of the near-field e vironment may act to sorb radionuclides, removing them fro.,e solution and retarding RT.

Oxides and oxyhydroxides of metals, such as Fe, Mn, and Si, are common fracture lining minerals in the YM system (Carlos, et al.1993) and may also be created by oxidation of materials introduced during the construction and operation of the repository (e.g., steel containers and rock bolts). Electrostatic sorption is a function of surface charge. Titration experiments with oxyhydroxides indicates that surface charge is a complex function of system chemistry, particularly pH (Davis and Kent 1990) For cations, such as UO/*, NpO,*, and Am 3*, oxyhydroxides exhibit a sharp sorption edge where, depending on radionuclides concentration and the number of available sites, sorption of cations increases from zero to nearly 100 percent over a relatively narrow pH range. In the presence of complexing ligands, such as COf , cation sorption typically decreases to zero with further increases in pH. For anions and oxyanions, such as SeO4 2 the reverse is true, sorption typically decreases in a gradual fashion with increasing pH (Davis and Kent,1990). Reactive surface areas can be high for the amorphous forms of oxyhydroxides suggesting the potentialimportance of these minerals as a sorbent phase. In addition, the potential for forming a sorptive oxide coating on less sorptive particles, such as quartz or feldspar, suggests an additional role for these minerals in radionuclides sorption (Robert and Terce,1989).

Other sorptive phases, such :.,s clays, occur at YlVI as secondary replacement products. In addition, clays may also develop as alteration products of vitrified waste and spent fuel.

Because of interlayer exchange sites and the large surface area resulting from their layered structure, clays can have a high cation exchange capacity. Smectite, vermiculite, and some kaolinite group clays expand upon interaction with water or organic fluids. This can change the interlayer spacing, and affect the degree to which radionuclides can penetrate the interlayer ion exchange sites and sorb onto clays (Goldberg, et al.,1991). Increasing ionic strength can reduce interlayer spacing, and ion exchange on planar sites is likely to be less. The edge sites (perpendicular to the silicate layers) also exhibit a surface charge that varies as a function of pH in a manner like that described above for oxides and oxyhydroxides. Actinide sorption on clays is pH-dependent (Zachara and McKinley,1993; Pabalan and Tumer,1997).

49

Zeolites, such as clinoptitolite, heulandite, and ana,-,,ie, are also likeiy to be important for retarding radionuclides at YM. Minerals, such as zeolites, exhibit a fixed charge developed by substitution of Al3* for Si"in the zeolite structure that is compensated by Na*, K', Ca 2*, and 2

Mg

• in the intracrystalline exchange sites. Sorption is typically by way of ion exchange in the intracrystalline sites (Davis and Kent.1990), particularly for the alkaline and alkaline earth elements, such as the short-lived radioisotopes of Cs* and Sr2 *, but there also appears to be a component of pH-dependent surface charge as well (Pabalan, et al.,1993; Pabaian and Turner, 1993).

At increasing temperature and high pH, calcite may be stable in the near-field environment (Murphy and Pabalan,1994). For radionuclides sorption, the surface charge of carbonate minerals is dominated by the balance between the dominant cation (Ca2 ' or Mg 2*) and the carbonate anion (CO32 ). For this reason, sorption on carbonates is a complex function of pH, solution chemistry, and p(CO2 ). Recent modeling efforts have focused on adapting surface complexation models to describe sorption at the interface between water and carbonate minerals (van Cappellen, et al., W93).

Mineral precipitation 9 dissolution can also affect the retardation of radionuclides migration due to introduction or removal of sorptive minerals. Minerals such as zeolites, clays, and oxides can be dissolved and reprecipitated (Bish,1993; Murphy, et al.1996), depending on temperature and fluid chemistry. In addition, removal of radionuclides from solution either due to the precipitation of stoichiometnc radioelement compounds or coprecipitation as an impunty in other minerals. is also affected by the temperature and chemistry of the sciution (Murphy and Prikryl,1996). Walton, et al. (1985) demonstrated this reaction experimental!y by machining circular flow channels into granite blocks and constructing thermal-convection loops to study the effects of heat and mass transport on radionuclides migration. A 40 C temperature gradient was applied across the system. Several radionuclides ('2sSb, "Co, and "Mn) were concentrated at the hot side of the experiment, probably by sorption on Fe oxyhydroxides. j

'"Ce and "Tc were present in elevated concentrations on the cold side of the apparatus {

Most sorption experiments are run at room temperature (20 to 30 C), and the effects of  ;

elevated temperature are poorly understood. Machesky, et al. (1994) indicate that the zero-point-of-charge (phac) of rutile decreases with increasing temperature. This change suggests i that negative charge development is enhanced for oxyhydroxides with increasing temperature, and that the pH edge for cation sorption would rrove to lower pH values at higher %mperatures.

Limited caten aata .'vr temperatures up to 85 C suggeal that sorption coefficierus for Am, Ba, Ce, Cs, Eu, Pu, Sr, and U on crushed tuff materials either remain constant or increase with increasing temperature (Meijer,1990). This is also the assumption made in current DOE TSPA transport models (TRW Environmental Safety Systems, Inc.,1995). However, there is a lack of j sample characterization before and after sorption, and large experimental uncertainties persist. {

These uncertainties and the empirical nature of the limited data make it difficult to extrapolate j over ranges in physical and chemical conditions likely in the near field.

Diffusion in addition to movement of a radionuclides-bearing fluid in response to system gradients in temperature, pressure, and chemistry, solutes can also move in and out of the fluid parcel 50 L

l l

l 1

1

[ _ _____________i

I". ,

(Manteufel, et al.,1993). These migration processes incluov. thermal or Soret diffusion, in response to the temperature gradient; ultrafiltration or reverse osmosis, in response to pressure

, gradients; and mass diffusion, in response to concentration gradients.

l Chemical osmosis (the movement of water in response to concentration gradients) has also been suggested by Carnahan (1984) as a major factor in fluid flux. Numerical simulations i

indicate that water flux in hydrologically-saturated clays from less to more concentrated zones can be as much as 3 to 4 orders of magnitude greater than the Darcy flux. This is counterbalanced by the flux of solute back across the membrane by pressure gradients (ultrafiltration or reverse osmosis). In coupled systems, Carnahan (1984) indicates that reverse osmosis is of minor significance compared to chemical osmosis.

A retardation mechanism that is potentially important at YM is migration of water from fractures, where transport may be relatively rapid, into the matrix where flow and transport are slow. Field studies at the Nopal i U deposit in the Pe6a Blanca mining district, Chihuahua, Mexico, suggest that diffusion into the matrix is of Innited importance in U retardation (Pearcy, et ..,1995).

Instead, U transport in 'actured tuff appears to be dominated by fracture flow and precipitation of a suite of secondary uranyl minerals that proceeds with time through hydrated uranyl cides to uranyl silicates and adsorption onto Fe oxyhydroxides and clays. Similar paragenesis have been observed in long-term dnp expenments using water, related to that from the J-13 well at YM, and unirradiated UO; (Wronkiewicz. et al ,1992).

Gas Transport Boiling would partition "CO2 into the gas phase, enharcing gaseous RT. Increased pH, perhaps through interaction with human-introduced materials in the near field, could result in increased partitioning of "CO2 into the liquid phase. Codell and Murphy (1992) performed one-dimensional simulations of "C transport in unsaturated rock. The results indicated that, after an early initial release of "C to the gas phase CO 2will dissolve into the aqueous phase and calcite precipitation will serve to sequester "C at longer times. The amount of gas transport is also sensitive to the thermalload imposed by the repository. Higher thermalloads cause venting of gas at the surface in numerical simulations (Light, et al.,1989). The calculated releases were also dependent on the travel time to the surface, which was, in turn, dependent on the Darcy velocity and the partitioning coefficient betweer the gaseous and aqueous phases.

Current DOE TSPA r., Web do not explicitly i,1c ude "Cu2 gas transport (i RW t:r*onmental Safety Systems, Inc.,1995). The decision by DOE to not include this mode of transport was based in part on recent recommendations of the National Academy of Sciences (National Research Council,1995), where it is considered that "CO2 release at the accessible ,

environment will be sufficien.tly diluted through mixing in the atmosphere to pose negligible  !

individual risk. Other potential gas phase species, such as i291 and "Cl, are assumed in some TSPA-95 scenarios to be transported as gases without any retardation through the EBS, and then to be dissolved in the aqueous phase (TRW Environmental Safety Systems, Inc.,1995).  ;

51 <

Colloid Transport Colloids involving radionuclides are typically called radiocolloids and have been divided into two types (Maiti, et al.,1989). "Tru? or "real" colloids are generally formed from hydrolysis, polymerization, condensation, or precipitation of radionuclides compounds in solution. True colloid stabilization is favored under alkaline conditions, especially in the case of highly-charged, redox-sensitive, species such as actinides (Maiti, et al.,1989; Choppin and Mathur, 1991). Olofsson, et al. (1982a,b) indicated that the formation of colloids is favored for the actir ides in lower (+3, +4) valence states. In the near field, where radionuclides concentrations can be relatively high, there is the potential for locally reducing-conditions, and the formation of true colloids could be favored.

In contrast to true colloids, pseudocolloids are formed when the radioelement sorb on small particles already present in the groundwater. In the near field, these particles may be either natural or introduced by human activity, and include organic and inorganic C, silica. clay particles, and oxyhydroxide compounds of metals, such as Fe, Mn, and Al. The presence, stability, composition, and sorptive capacity of these particles are dependent on different aspects of the chemistry of the groundwater system, including pH, Eh, ionic strength, and p(CO2). Further complicating the behavior of pseudocolloids is ihe possibility of nonsorptive particles being coated with sorptive materials (Robert and Terce,1989). Experimental evidence has also demonstrated that colloids can be formed as secondary precipitates and clay alteration products and released from HLW forms (Bates, et al.,1992, Ebert and Bates,1992: Fmn et al.,

1994a).

Colloid Stability if actinides and other radioelement can sorb onto pseudocolloids or form true colloids, the stability of the particles in suspension is of criticalimportance in colloid-mediated transport. In the case of pseudocolloids, the availabikty of natural or human-introduced particles for RT in solution is affected through several mechanisms. Particles may be introduced through the nucleation and growth of crystalline and amorphous mineral phases in response to chemical supersaturation. in a groundwater system particles may also be released from the aquifer  ;

matrix to solution. This release may be due to electrostatic dispersion, brought about by ,

changes in solution ionic strength; mechanical disruption of primary and secondary minerals, owing to shearing and poding of mineral surfaces by hydrodynamic forces or autogrinding between suspended pertcles; or dissolution of a more soluble matrix to expose and reiease the colloid. Human activity associated with a HLW repository may also introduce colloidal materials in the form of organics associated with dissolution of vitrified waste-forms, and organic matter used in drilling, construction, and repository operations (Travis and Nuttall,1985). Corrosion products from the EBS may also serve as sources for pseudocolloid formation.

For metal oxyhydroxides, particle stability is a function of pH, Eh, particle size, and the tota!

concentration of the metal (e.g., Fe, Mn, Al, Ti, Si) in solution. The presence of other ligands, such as HCOi and SO 2 , and p(CO 2), can also affect the formation of oxides by consuming metal ions in the precipitation of carbonate and sulfate solids. If changes in solution chemistry result in desorption of radioelement, they are free to sorb onto the immobile medium, in this case, colloid transport becomes less of an issue for PA.

52 f

l

l In addition to pH controls on the sorptive' capacity of colloidal particles, the stability of the colloidal suspension of charged particles varies as a function of ionic strength, solution chemistry, and pH. For example, at low ionic strengths, the electrostatic double-layer (EDL) expands outward from particle surfaces stabilizing the colloids in solution through electrostatic repulsion. At higher ionic strengths the double layer collapses, and the charged particles begin l to flocculate (agglomerate) and come out of suspension due to gravity settling and filtration.

Variations in overall solution chemistry and moisture content of the medium influence the raagnitude of the ionic strength effect. Colloidal stability also depends on the predominant ion (s) in solution. For example, the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory that considers van der Waals attraction and electrostatic repulsion can be used to calculate the critical coagulation concentration (CCC) at which colloids are just destabilized in electrolyte solutions. DLVO theory predicts that the CCC should decrease with increasing valence, in proportion to the ratio 1/z*, where z represents valence, in agreement with the empirical relationship expressed in the Schulze-Hardy rule (Hunter,1987). Variations in pH also affect i particle stability. At low pH. the positive surface charge of variably-charged surfaces, such as clay edge sites and oxyhydroxides, is high. For clays, these conditions result in increased bonding of positively-charged crystalline edges to negatively-charged planar sites. Positively charged oxides will also bond to negatively charged clay surfaces and organic macromolecule (Ryan and Gschwend,1990). Under these conditions, clay dispersion is low, flocculation and a90l omeration occurs, and the suspension is destabilized. As the pH increases towards the pHs the positive surface charge of the oxides decreases and bonding to clays diminishes. At high pH, edge sites and oxyhydroxides exhibit a negative surface charge and actively repel the negatively-charged clays. dispersion is enhanced. and the clay colloids are kept in suspension (Suarez, et al.,1984).

Colloid Transport and Filtration The effectiveness of colloids in enhancing or retarding radionuclides migration is dependent on the efficiency of particle transport through the groundwater system. Colloid migration may be (nhanced relative to fluid flow due to volume exclusion effects and reduced interaction between ,

the particle and medium. Conversely, colloids rnay be ratarded through various physical and chemical filtration mechanisms resulting from interaction between the different phases of the colloid-rock-water system. Based on filtration tneory, Herzig, et al. (1970) described the transport of colloids in suspension using a modification of the convection-dispersion equation.

Empirical parameters are typically designed to ref'ect the efficiency with which a porous mediurr 'ilters partWas from suspension, and porosity is arfected by particle der mion.

McDowell-Boyer, et al. (1986) divided filtration processes into three basic classes: (i) surface (cake) filtration' (ii) straining,' and (iii) physical-chemical filtration. Surface filtration involves building a barrier at the interface between the water and pore. This type of filtration occurs when the particles are too large to enter the pores of the medium. As the particles are stopped at the surface, they are held in place by the fluid flow, and a mat or cake is gradually formed.

With time the filter cake thickens and its porosity and permeability decrease through compression. Fluid flow through the mat decreases, and there is a pressure drop across the cake. Filter-cake permeability is a!so a function of padicle aggregation. Destabilized colloidal suspensions (e.g., high ionic strength) tend to form a more porous arrangement than those cakes formed from highly-dispersed stable suspensions (McDowell-Boyer, et al.,1986).

53 1

If the particles are small enough to enter the porous medium, the tortuous path they must follow may eventually lead to a constriction that is too small for them to pass through. This leads to a straining of the colloids from solution. For example, in an attempt to consider a population of suspended particles with variable diameter, Sherard et al. (1984) observed that fine particles would not enter a porous medium for the value of (dm,15)/(dp,85) < 9 where (dm,15) exceeds the pore diameter of 15 percent by weight of the coarse media, and (dp,85)is the upper diameter limit for 85 percent of the suspended particles by weight.

Particles may be removed from suspension by interaction with the pore walls, either through physical processes such as Brownian diffusion, gravitational sedimentation, or through chemical processes such as sorption due to electrostatic attraction. Once particles have been deposited, there is the possibility that they may be resuspended. The distances calculated for the energy attachment well(0.3-1 nm) are generally smaller than the diameter of the particle. While London-van der Walls forces generally predominate at these ranges, energy provided from Born repulsive forces, or thermal and hydrodynamic energy, can overcome the attraction energy well and lead t, particle erosion and re-entrainment. An additional possibility is that a decrease in the solution ion,;c strength may evtend the EDL, leading to particle release (Kallay, et al.,1987). Kallay, . al. (1987) also indicate that sweeping the resuspended particle away from the surface is necessary to prevent reattachment.

Although the size of colloids makes them vulnerable to several different filtration mechanisms, it is also possible that particle size (Bales, et al.1989) wdl lead to volume exclusion and a less tortuous, more rapid path to the accessible environment. In pores and fractures, the water velocity distribution is such that the maximum velocity is along the centerline of the fracture, while minimum velocity occurc at the fracture wall. Because of their size, colloids can never "expenence the minimum water velocity and, as such. the average colloid velocity will be larger than that of the water. In general, this effect, called hydrodynamic chromatography (de Marsily, 1986), becomes more pronounced with increasing particle diameter. In addition, electrostatic repulsion associated with charged particles will tend to keep the particles away from the surfaces, further enhancing the effect. Since the particle charge is a function of pH and ionic strength as discussed above, hydrodynamic chromatography in a natural environment varies as a function of solution chemistry (de Marsily,1986).

Most of the research on colloid formation and transport is based on studies of saturated media.

Colloid transport through unse'urated media, su+ as at YM, is poorly understood. The presence of a gas ph.w: r"ay influence particle transport by particle attacament to the bubble surface (Wan and Wilson,1994).

Localized reducing conditions could be promoted by near-field hydrologic effects and phase variations. Local fluctuations of reducing and oxidizing conditions in the near field due to an unstable hydrologic regime could also induce secondary chemical effects such as the formation of colloids (Buddemeier and Hunt,1988; McCarthy and Zachara,1989).

4.4.2.2 Effects of Engineered Materials en Radionuclides Transport Staff analysis of the effects of engineered and man-introduced materials on RT assumes the reference case design for the proposed repository. The base case design assumes pre-cast j 54 l

concrete liners, rather than carbon steel ribbing, for drift support. The effects of engineered and man-introduced materials other that cementitious material are likely to be important to radionuclides release; however only a preliminary discussion is presented below. In addition, the following preliminary discussion of cementitious materials reflects only a limited effort by the ENFE KTl to address this topic during this FY. Thus, a more complete discussion of the effect of engineered materials on radionuclides transport will be presented in Revision 1 of this IRSR.

The principal organic components of natural soils and waters are humic materials (Choppin, 1988), and other organics may be introduced during repository construction and operation (e g.,

solvents, fuels, etc.). The anionic charge of organic molecules allows them to bind readily to cationic species in solution. For enmple, humic substances can complex ions in solution principally through oxygen donor sites, and can bind relatively highly-charged cations, such as heavy metals and transuranic radionuclides. Also, organic molecules may become bound as gels and coatings to the surface of inorganic particles, such as clays and oxides, changing the sorptive behavior to reflect the ornanic coating (Robert and Terce,1989). For example, stuc'ies of Kohler, et al. (1992) indicate that the presence of ethylenediaminetetraacetic acn sdDTA) in milhmolar concentrations (10" to 10 ' M) can sign 3cantly reduce the amount of Np(V) sorbed on kaolinite. EDTA concentrations of the order 10 6 M, however, have only a slight effcci on the sorption behavior.

Oxides and oxyhydroxides of metals such as Fe, Mn, and Si, found in many fractures of YM, may also be created by oxidation of materials introduced during the construction and operation of the repository (e.g , steel containers and rock bolts). These highly sorbent minerals are hkely to be irnportant to RT within the near field.

Radionuclides transport potentially can be affected by the presence of cementitious matenals in the near-field environment of the proposed repository. Cement hydration products proviae a -

multitude of sorption sites that could aid in retarding the migration of radionuclides (Atkins, et al.,1990,1991) from the EBS to the host rock In addition, as pointed out :n Section 4.3.2, the -

persistent alkaline pH (>10) of pore fluids in contact with hydrated cement phases favor precipitation of a wide variety of radionuclides. On the other hand, mineral alteration due to

alkaline solutions and precipitation of secondary phases could affect the sorptive and l retardation ability of ine geologic barrier, and could also affect its hydraulic properties (porosity l and permeabihty). For example, calculations t / Lichtner and Eikenberg (1995), using a geochemical Pansport model C.tPATH), indicatad that interaction between a hyperalkaune plume released from t ument-based low-level radioactive waste repository and a mar. ;,ost I rock resulted in a rapid decrease in porosity of i

e host rock several meters from the repository, due to precipitation of secondary phases, whereas porosity increased at the interface of the marl host rock and the cement due to mineral dissolution.

One of the strongest effects on RT is likely to be on the potential for colloidal transport. Recent work suggests that colloids that could be formed within the WP or from the spent fuel would most likely be agglomerated as a result of interaction with alkaline fluids associated with cementitious materials (Savage,1997). Since a concrete invert is envisioned for the drifts, any colloids generated hydrologically up gradient (i.e., in the WP) would be subject to alkaline pore fluids associated with the invert.

55 E _ _ _ _ _ _ _ _ _ _

A potentially adverse scenario is that of flow of a hyperalkaline fluid along fractures in the tuffaceous host rock of the proposed repository. Dissolution of the tuff could lead to widening of the fractures and enhancement of groundwater flow and RT. Alternatively, precipitation of calcite and CSH phases along the fracture and matrix interface could seal the fractures from the matrix, producing isolated channels through which RT could occur relatively unimpeded by matrix diffusion. However, if sufficient amounts of calcite and CSH phases are precipitated along fracture walls, reduction in fracture porosity and permeability, or fracture plugging, could result in diminished flow and transport through the repository. Preliminary calculations by Lichtner, et al. (1997) suggest that strong alteration of the YM tuff host rock and of cement in contact with the tuff could result from interaction of cement and tuff pore waters and the respective minerals. Further analyses are required to develop a better understanding of the effect of alkaline plume migration on RT.

4.4.2.3 Radiolysis Effects on Radionuclides Transport Experiments on spent fum leaching withcat imposed irradiation (Finn et al.,1994a,b) are representative of potential autoradiolytic effects from spent fuel alpha radiation. These experiments imply that nascent hydrogen, H', plays a role in reducing carbonate in solution tu formate and oxalate. During RT, the coexisting reduced and oxidized species could become separated, leading either to a net reduction or net oxidation of the environment where radionuc; ides are concentrated. Furthermore. it is possible that reduced U'* may form mobile complexes with the formate ano oxalate radiclysis products (Finn, et al ,1994a,b).

Naturally occurnng U'* in sedimentary rocks is commonly correlated with organic matter (Pierce, et al. 1955,1964; Nash, et al.,1981), but the process by which the association arises is not fully understood. Uranium is readily transported in the uranyl (UO 22*) state as carbonate complexes. Uranyl adsorption on organic material containing oxygen-bearing functional groups and as -COOH, -COO, and -OH is favored and probably represents the first step of U mineralization. Subsequent reduction of the adsorbed uranylion by the organic matter or other reducing species, and eventual precipitation of a U'* mineral (e g., uraninite or coffinite),

follows. It appears, however, that the organic material hosting the U sometimes accumulates from solution in the form of asphaltite- or thucholite-type nodules, (e.g., Pierce, et al.,1955).

The growth of these nodules could be an indication of autogenous radiolysis, during which water-miscible hydrocarbons are scissioned by radiation and condense.

Alternatively, this process nignt be inhibited by toe reduction of bicarbonate to formate cr acetate. Thc reactive H' could reduce adsorbed uranyl complexes to uraninite. As more UO,  !

precipitates, a chemical potential gradient in UO22* carbonate complexes would be set up that would diffuse toward the precipitated UO 2, thereby, increasing the probability of U adsorption, reduction, and precipitation. Further study of U coprecipitation with organic material is required to establish the validity of these hypotheses.

Although both oxidizing and reducing radiolytic effects on waste forms can be hypothesized, the preponderance of evidence suggests that oxidation (and possibly acidification) will be dominant over reduction. Nevertheless, the potential for radiolytic reduction needs to be considered and i its effect on mobil.ty understood. Notably, DOE's TSPA completed in 1995 (TRW {

Environmental Safety Systems, Inc.,1995) does not consider potential radiolytic effects on the

]

l 56 )

d I i 1

)

[- -

l l

I source term cr RT. The preceding discussion dernonstrates that quantification of radiolytic effects is fraught with uncertainty. Nevertheless, it should be possible to quantitatively calculate ranges of possible states. Once this is accomplished, perhaps radiolysis can be incorporated into PA models by ensuring that probability distribution functions for parameters, such as solubility, sorption coefficient, and release rate, cover the ranges of possible effects.

4.4.2.4 Microbial Effects on Radionuclides Transport l This information will be provided in Revision 1 of this IRSR, currently scheduled for FY 1998.

l l

l 57

5.0 STATUS OF ISSUE RESOLUTION AT THE STAFF LEVEL NRC staff has re.ised concerns about the U.S. Department of Energy's (DOE) site characterization and performance assessment (PA) programs in areas related to evolution of the near-field environment (ENFE). These concerns have been documented in the following:

• Staff Site Characterization Analysis (SCA) of the DOE's Site Characterization Plan (SCP), Yucca Mountain Site, Nevada (U.S. Nuclear Regulatory Commission,1989);

• the NRC audit review of DOE's 1995 total system performance assessment (TSPA-95; U.S. Nuclear Regulatory Commission,1996b); and

• a letter from NRC to DOE with findings of the staff review of DOE's thermohydrology testing and modeling program (U.S. Nuclear Regulatory Commission,1997).

No new open items have heen identified in this isst.e resolution status report (IRS".) on the evolution of the near ' ld environment. A discuscion of the findings from each of the sources listed above is provided below. Consideration with regard to the status of the items, and whether they are resolved or open at the staff level, is discussed.

5.1 U.S. Nuclear Regulatory Commission Review of U.S. Department of Energy Site Characterization Plan The NRC review of the DOE SCP (U.S. Nuclear Regulatory Commission,1989) resulted in eight comments and one question related to the ENFE key technicalissue (KTl). Since the time of the SCA, DOE has adopted a revised program plan (U S. Department of Energy,1996b) and the waste isolation and containment strategy (WCIS). The WCIS developed a number of hypotheses concerning isolation attributes of the proposed high-level radioactive waste repository (U.S Department of Energy,1996a). DOE's refocussed program, a result of Congressional direction (U.S. Department of Energy,1996b; page 11), incorporates the WCIS and focuses on the remaining technical questions that have been demonstrated, through TSPAs, to be important to waste containment and isolation. As a result of the refocused program, many of the study plans have changed in scope, beer' deferred, or canceled (U.S.

Department of Energy,1997a Appendices A and G). Staff will continue to track DOE progress on the technical concerns incorporated in the SCA comments and question as ps. of ;he issue resolution process of the ENFE KTI. As part of that effort, a discussion of the SCA comments and question, and their relationship to the refocused DOE and NRC programs, is provided below.

• Comment 25 (U.S. Nuclear Regulatory Commission,1989; page 4-29). The SCP does not provide the rationale for additional testing to obtain information on the effects of package degradation products and interactions between and among radionuclides on sorption.

Comment 25 relates to subissue 3 (rate of release of radionuclides from breached waste packages (WPs)] and subissue 4 [ radionuclides transport (RT) through engineered barriers and 58

I, ,

}

natural barriers). While the Container Life and Source Ten 'CLST) KTl has lead responsibility for this comment, the ENFE KTl also considers the comment important to issue resolution.

Staff considered this comment open based on the decision that the DOE commitment to study the effects of microbial activity in the near-field environment, in itself, was not sufficient reason to resolve comment (U.S. Nuclear Regulatory Commission,1991) Comment 25 raised two issues: effects of WP degradation products on waste mobilization; and interaction among and between radionuclides on sorption. Based on the low solubilities of radionuclides, the staff

' considers interaction between and among radionuclides insignificant in sorption processes.

With regard to the effects of degradation products, the staff still considers this topic important. i NRC technical concerns are embodied in DOE :: WCIS hypotheses 9 and 10. Staff, through its focused review of the evolving DOE program, will track progress in DOE's characterization of the effects of waste package degradation products on waste mobilization as part of the issue I resolution process for the ENFE KTI.

e Comment 29 (U.S. Nuclear Regulatory Commission,1989; page 4-32). This comment concerns activities to evaluate the effects of radioactive decay heat, th; ...a, lear radiation field, and the effect of non-s;te soecific microorganisms introduced during site construction.

Comment 29 relates to subissue 2 (WP lifetime), subissue 3 (rate of release of radionuclides from breached WPs), and subissue 4 (RT through engineered barriers and natural bamers).

Staff considered this comment resolved based on DOE's commitment to study the effects of microbial activity in the near-field environment (U.S. Nuclear Regulatory Commission,1991).

DOE stated that this work would be covered in the Study 8.3.4.2 4.1 (Characterization of Chemical and Mineralogical Changes in the Post-Emplacement Environment). NRC technical concerns are embodied in DOE's WCIS hypothesis 9. The original comment is considered resolved, however the staff, through its focused review of the evolving DOE program, will track progress in DOE's characterization of microbial effects as part of the issue resolution process for the ENFE KTI.

e Comment 79 (U.S. Nuclear Regulatory Commission,1989; page 4-66). It has not been demonstrated that the test environment in waste package corrosion tests is fully representative of the repository environment.

Comment 79 relates 5 subissue 2 (WP lifetim 4 The ENFE KTl has lead responsibility for this comment, and the CL6 i' r;TI also considers t, ; comment important to issue resn'" tion. Staff considered this comment open because DOE indicated that the test environments for WP corrosion tests will evolve as site data and detailed designs become available (U.S. Nuclear Regulatory Commission,1991). NRC technical concerns are embodied in DOE's WCIS hypotheses 6,7, and 8. As part of the issue resolution process for the ENFE KTl, staff will track progress in DOE's WP corrosion tests to ensure that test environments are fully representative of the repository environment.

e Comment 81 (U.S. Nuclear Regulatory Commission,1989; page 4-67). Investigations into the stress corrosion cracking (SCC) behavior of the container alloys assume that the container surface will be either homogeneously dry or homogeneously wet, but in 59

i i~

the corrosion model (7.4.5.4.6), it is stated that "the waste pack + will most likely not be uniformly wet."

Comment 81 relates to subissue 2 (WP lifetime). The CLST KTl has lead responsibility for this comment, and the ENFE KTl also considers the comment important to issue resolution. This comment was open because DOE indicated that the test environments for WP corrosion tasts will evolve as site data and detailed designs become available (U.S. Nuclear Regulatory Commission,1991). Subsequently, tests involving dripping onto containers have been

performed. NRC technical concerns are embodied in DOE's WCIS hypotheses 6,7, and 8.

l The staff, through its focused review of the evolving DOE program, will track progress in DOE's characterization of the SCC behavior of the container alloys, as influenced by coupled processes, as part of the issue resolution process for the ENFE KTI.

e Comment 84 (U.S. Nuclear Regulatory Commission,1989; page 4-68). The issue resolution strategies and testing programs for design of the WP and the ennineered barrier system (EBS) do not lake into account the full range of reasonable imely natural l conditions (" anticipated processes end events") that, with current understanding at the site, might be expected to affect performance of these barriers.

Comment 84 relates to subissue 2 (WP lifetime), subissue 3 (rate of radionuclides release from l breached WPs), and subissue 4 (RT through engineered and natural barriers). The CLST KTl l has lead responsibility for this comment, and the ENFE KTl also considers the comment

! important to issue resolution Staff considered comment 84 open since the tests and analyses

! did not reflect the fud range of potential anticipated processes and events and. as need be, unanticipated processes and events (U.S. Nuclear Regulatory Commission,1991). NRC technical concerns are embodied in DOE's WCIS hypotheses 6,7,8,9, and 10 The staff will track progress of DOE's testing prcgrams for design of the WP and the EBS. The staff will follow this effort as part of the issue resolution process for the ENFE KTI.

l j e Comment 89 (U.S. Nuclear Regulatory Commission,1989; page 4-71). Grouts,

! cements, and organic materials used in the repository may change the local pH of the repository and affect corrosion of the metal waste containers and the local teach rates of radionuclides from the glass.

Commer.t 89 relates to rbissue 2 (WP lifetNe), subissJe 3 (rates of radionuclides releam from breached WPs), and sucissue 4 (RT through er,aineered and natural barriers). The ENFE KTl has lead responsibility for this comment, and the CLST KTl also considers the comment important to issue resolution. Staff considered comment 89 open based on DOE's response.

I Its response indicated that testing programs will investigate how water chemistry is changed by l' the waste package, and other repository materials, and how such changes affect the corrosion l

of the containers and the leaching of radionuclides. However, no details were provided (U.S.

Nuclear Regulatory Commission,1991). NRC technical concerns are embodied in DOE's WCIS hypotheses 6,7,8, and 9. Through its focused review of the evolving DOE program, staff will track how DOE considers the impact of grouts, cements, and organic materials on corrosion of waste containers, and leach rates of radionuclides, as part of the issue resolution process for the ENFE KTI.

60 e

e

~

1

• Comment 90 (U.S. Nuclear Regulatory Commission,1989; page 4-71). The effects of I varying oxygen concentrations on corrosion of the metal canisters are not considered-Comment 90 is related to subissue 2 (WP lifetime). The CLST KTl has lead responsibility for this comment, and the ENFE KTl also considers the comment important to issue resolution.

l This comment was considered open based on the response of DOE which indicated that the effects of varying oxygen concentration on the corrosion of the metal container will be considerec when site data detailed designs and performance scenarios are available (U.S.

Nuclear Regulatory Commission,1991). NRC technical mncerns are embodied in DOE's WCIS hypotheses 6,7, and 8. Through staff's focused review of the evolving DOE program, how DOE evaluates the effects of vary ng oxygen concentrations, as they are influenced by coupled processes, on corrosion of WPs will be tracked as part of the issue resolution process for the ENFE KTI.

• Comment 92 (U R Nuclear Regulatory Commission,1989; page 4-72). The approach for delineation of the boundary of the disturbed zone does not include all physical or chemical propertes which will have changed as a result of heat generated by the emplaced radioactive wastes such that the resultant change of properties may have a significant effect on the performance of the geologic repository.

Comment 92 relates to all of the subissues of the ENFE KTl Staff considered this comment resolved based on the DOE response which indicates Activity 1.6.5 2 Definition of the Disturbed Zore will reevaluate and if necessary refine the boundary of the disturbed zone. This ongoir.g activity may be a result of changes in NRC guidance and in DOE understanding of repository property effects (U.S. Nuclear Regulatory Comm:ssion 1991). NRC technical ccacerns are embodieo in DOE's WCIS hypotheses 2, 3,4,6,7 9, and 10. The original commerit is considered resolved. However the staff, through its focused review of the evolving DOE program, will track progress in DOE's characterization of the evolution of the near-field environment and its impact on performance as part of the issue resolution process within the ENFE KTI.

• Question 30 (U.S. Nuclear Regulatory Commission,1989; page 4-115). It is stated that the expected quality of the water is such that it will have !ittle impact on the long-term integrity of the WPs. What is the expected quality of the water and how might this quality vary over tha hfetime of the repository?

Question 30 is related to subissue 2 (WP hfetime), subissue 3 (rates of radionuclides release from breached WPs), and subissue 4 (RT through engineered and natural barriers). The ENFE KTl has lead responsibility for this question, and the CLST KTl also considers the question important to issue resolution. Staff considered this question open because DOE indicated that the expected water quality will be unknown until the results of activities in Study Plan 8.3.4.2.4.1 are completed. NRC technical concerns are embodied in DOE's WCIS hypotheses 6,7, and 8.

Staff will track progress in DOE's evaluation of water quality and its impact on performance as part of the issue resolution process within the ENFE KTI.

61

5.2 U.S. Nuclear Regulatory Commission Aw.t Review of U.S. Department of Energy TSPA-95 The environment of the near field can affect the modes of corrosion attacking WP surfaces.

l The NRC audit review of TSPA-95 (U.S. Nuclear Regulatory Commission,1996b) identified that several potential WP failure modes were not considered and, as a result, the calculations may be nonconservative. Staff concerns related to DOE's hurnid air corrosion model were raised because the model does not account for the effect of aggressive groundwater, the hygroscopic nature of corrosion products, or other forms of capillary condensation in the pores of oxide scale formed on the container surface. The staff also considered the assumed distribution of pitting factors to be nonconservative and lacking a mechanistic basis. Staff also qucioncd whether episodic wetting and drying might increase the corrosion rate. These comments and questions relate to subissue 2 (the effects of coupleo processes on WP lifetime). This concern remains open but could he addressed by conducting sensitivity analyses to determine the impact of corrosion modes on containment As a general comment, it was noted that possib:e geochemical variations in the near-field environmental cond' is are considered reasonzbly well in the TSPA-95 (U.S. Nuclear Regulatory Commission,1996b). However, minimal effects of changes in the geochemical

- environment are employed in the performance calculations. This comment relates to subissue 3 (the effects of coupled processes on rates of radionuclides release from breached WPs) and subissue 4 (the effects of coupled processes on RT through engineered and natural barriers).

l Sensitivity analyses using codes capable of handling coupled processes can be implemented to 1 address the effect of near-field chemistry on the solubility of radionuclides. Given the empirical l relationships for solubilities of Np, Pu, and Am as a f Jnction of pH and temperature from TSPA-95, sampling on pH would result in concentrations of these radionuclides responding to changes in near-field chemistry in a reasonable manner.

5.3 U.S. Nuclear Regulatory Commission Review of U.S. Department of Energy Thermal Modeling and Testing Program Staff recently reviewed DOE's Thermohydrology Testing and Modeling Program and transmitted the findings (U S. Nuclear Regulatory Commission,1997). The comments were directed at the effects of the thermal perturbation on flow in the near field. These comments relate to subissue 1 (the effects of coupled procmses on the rate of seepage into the repository), but also, L, th e other subissues in that water flow to and from the waste packages is important to WP lifetime (subissue 2), rates of radionuclides release from breached WPs (subissue 3), and RT through engineered and natural barriers (subissue 4).

Three comments were submitted to DOE. The first comment dealt with its thermal testing i strategy. A concern exists that an accelerated drift-scale heater test, at thermalleads higher than those expected at the repository, would possibly mask potentially important heat and mass transfer phenomena that might be present dunng the operation of a high-level waste repository.

The second comment raised the issue that the application of an equivalent continuum model approach, or alternative approaches, to modeling liquid flow to the waste packages has not been demonstrated. The last comment concerned the impact of thermal-hydrologic-chemical 62

(THC) coupling on flow and waste package containment (suoissue 1 and subissue 2 of this j KTl). - i l

l- DOE's response (U.S. Department of Energy,1997b) to the NRC letter addressed the staff's concern regarding conceptual models describing heat and mass transfer in the near-field environment. DOE noted that dual permeability models (DKM) and stochastic DKM conceptual models have undergone considerable refinement and development and are being implemented in at least the single-heater test analysis.

l DOE rerponded to the comment on THC coupling by indicating significant progress has been i achieved in modeling of THC coupled effects They cited several recent publications that document this progress. These documents and citations provided support for all three responses to the NRC comments. These citations are currently under review.

5.4 Status of issue Resolution at the Staff Level l

l As noted in section 5.1 of this report, two items reliting to the ENFE KTI, icaulting rrom the staff review of the DOE SCP, have been previously resolved (comments 29 and 92). The rcrno;ning ENFE-related items from the review of the DOE SCP (comments 25,79,81,84,89, and 90; question 30) are considered to be open. The staff concerns noted in the audit review of TSPA-95 and the DOE's response to the NRC comments on their thermohydrology testing and modeling program remain to be addressed These comments and the technical concerns embodied in the SCP comments and question remain part of the NRC's refocused relicensing program within the ENFE KTl and will be addressed as part of the issue resolution process.

l 1

i 63

_ _ _ _ = - _ - _ - _ _ _ _ _ A

- +

6 REFERENCES Ames, L.L., Jr., "Some zeolite equilibria with alkali metal cations," American Mir,eralogist, Vol.

49,pp.127-145.1964.

. Angell, P., et al., " Evolution of the Near-Field Environment in the Proposed High-Level Waste Reposito,; at Yucca Mountain-A Review of Hypotheses," Prepared for the U.S. Nuclear Regulatory Commission by the Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas,1996.

Apted. M J., " Natural analogs for the predictive reliability of the engineered barrier system for high level waste," in B.I. Come and N.A. Chapman (eds.), Fourth Natural Analogue Working Group Meeting and Pogos de Caldas Project, EUR 13014 EN, Commission of the European Communities, Luxembourg,1990 Arthur, R.C , and W.M. Murphy, "An analysis of gas-water-rock interactions during boiling in partially saturated tuff," Sciences Gsologiques Bulletin, Vol. 42, pp. 313-327,1989.

Atkins, M . A.N. Beckley, and F.P. Glasser, " Influence of cement on the near-field environment and its specific interactions with uranium and todine." Radiochimica Acta Vol. 44/45, pp.

255-261,1988.

Atkins. M., et al., " Assessment of the performance of cement based comoosite material for radioactive waste immobilization." m V.M. Oversby and P.W. Brown (eds.), Scientific Basis for Nuclear Waste Management X///, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 176, pp.117-127,1990.

Atkins, M , F P. Glasser and A. Kindness, " Phase relations and solubility modeling in the CaO-SiO,-Al;O3-MgO-SO3-H3 0 system: For application to blended cements,"in T. Abrajano, Jr., and L.H. Johnson (eds.), Scientific Basis for Nuclear Waste Management XIV, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 212, pp. 387-394,1991.

Atkins, M., D. Damidot, and F.P. Glasser, " Performance of cementitious systems in the repository," in A. Barkatt and R./.. Van Kony.x iburg (ads.), Scientific Basis for ^/uclear Waste Management XV//, Matenalt, desearch Society, httsburgh, Pennsylvania, Symposium Proceedings 333, pp. 315-326,1994.

- Atkinson, A., N.M. Everitt, and R.M. Guppy, " Time dependence of pH in a cementitious repository," in W. Lutze and R.C. Ewing (eds.), Scientific Basis for Nuclear Waste Management XII, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 127, pp.

439-446,1989.

Baca, R.G., and M.S. Jarzemba, " Detailed Review of Selected Aspects of Total System Performance Assessment-1995." Al 5708-761-711, Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas,1997.

64 5

Bales, R.C., et al.," Bacteriophage transport in sandy soil and fractured tuff," Applied and Environmental Microbiology, Vol. 55, pp. 2,061-2,067,1989.

4 Bates, J.K., et al,., " Colloid formation during waste form reaction: Implications for nuclear waste disposal," Science, Vol. 256, pp. 649-651,1992.

I Bates, J.K., et al., " Yucca Mountain Project-Argonne National Laboratory, Annual Progress l Report, FY 1994," ANL-94/42, Argonne National Laboratory, Argonne, Illinois,1995.

Bish, D.L., " Smectite Dehydration and Stability. Applications to Radioactive Waste Isolation at )4 Yucca Mountain, Nevada," LA-11023-MS, Los Alamos National Laboratory, Los Alamos, New Mexico,1988.

Bish, D.L., " Evaluation of Past and Future Alterations in Tuff at Yucca Mountain, Nevada, Based on the Clay Mineralogy of Drill Cores USW G-1, G-2, and G-3," LA-10667-MS, Los Alarr os National Laboratory I as Alamos, New Mex:co,1989.

l l

Bish, D.L., " Study Plan for Kinetics and Thermodynamics of Mineral Evolution at Yucca Mountain, Nevada." YMP-LANL-SP 8.3.1.3 3.2, Los Alamos National Laboratory, Los Alamos, New Mexico,1993.

Bish, D.L., and S J. Chipera, ' Revised Mineralogic Summary of Yucca Mountain, Nevada,"

i.A-11497-MS. Los Alamos National Laboratory, Los Alamos, New Mexico 1989 i Bish. D.L., and J.L. Aronson, 'Paleogeothermal and paleohydrologic conditions in silicic tuff from Yucca Mountain. Nevada," Clays and Clay Minerals, Vol. 41(2), pp.148-161,1993.

Bish, D.L , et al., " Summary and Synthesis Report on Mineralogy and Petrology Studies for the Yucca Mountain Site Characterization Project," WBS Element 1.2.3.2.1.1 and 1.2 3.2.1.2, Studies 8.3.1.3.21, 8.3.1.3.2.2, and 8.3.1.3.3.2, Los Alamos National Laboratory, Los Alamos, New Mexico,1996.

Blesa, M.A., P.J. Morando, and A.E. Regazzoni, Chemical Dissolution of Metal Oxides, CRC Press, Boca Raton, Florida,1994. i Bowers, T.S., and R.G. Burr % " Activity diagram:, for clinoptilolite: Susceptibility of this zeolite to further digenetic reactions," American Mineralogist, Vct. 75, pp. 601-619,1990.

Bradbury, M.H. and F.-A. Sarott," Sorption Databases for the Cementitious Near-Field of a L/lLW Repository for Performance Assessment," PSI Bericht Nr. 95-06, Paul Scherrer Institute, Wurenlingen, Switzerland,1995.

Buchwald, V.F., and R.S. Clarke," Corrosion of Fe-Ni alloys by Cl-containing akaganeite (B-FeOOH): The Antarctic meteorite case," American Mineralogist, Vol. 74, pp. 656-667,1989.

i 65 )

)

Buddemeier, R.W., and J.R. Hunt, " Transport of conoidal contaminants in groundwater:

Radionuclides migration at the Nevada Test Site." Applied Geochemistry, Vol. 3, pp. 535-548.

1988.

Byers, F.M., Jr., " Petrochemical Variation of Topopah Spring Tuff Matrix with Depth (Stratigraphic Level), Drill Hole USW G-4, Yuc,ca Mountain, Nevada " LA-10561-MS, Los Alamos National Laboratory, Los. Alamos, New Mexico,1985.

Carlos, B.," Fracture-Costing Minerals in the Topopah Spring Member and Upper Tuff of Calico Hills from Drill Hole J-13," LA-11504-MS, Los Alamos National Laboratory, Los Alamos, New Mexico,1989.

Carlos, B.A., D.L. Bish, and S.J. Chipera,. " Fracture-lining minerals in the lower Topopah spring tuff at Yucca Mountain," in Proceedings of the Second Intemational Conference on High Level Radioactive Waste Management, American Nuclear Society, La Grange Park, Illinois, pp.

486-493,1991.

Carlos, B.A., et al., ,acture-lining manganese oxide minerals in silicic tuff, Yucca Mountain, Nevada," Chemical Geology, Vol.107, pp. 47-69.1993.

Carnahan. C L.," Thermodynamic coupling of heat and matter flow in near-field regions of waste repositories," in G.L. McVay (ed.), Scientific Basis for Nuclear Waste Management Vll, Materials Researen Society, Pittsburgh, Pennsylvania. Symposium Proceedings 127. pp.

1 023-1,030,1984.

Chan, K.S., "A micromechanical model for predicting hydnde embnttlement in nuclear fuel cladding material," Jcumal of Nuclear Materials, Vol 227, pp. 220-236,1996.

Chigira, M. and M. Watanabe, " Silica precipitation behavior in a flow field with negative temperature gradients," Joumal of Geophysical Research, Vol. B8(99), pp.15,539-15,548.

1994.

Choppin, G.R., " Chemistry of actinides in the environment," Radiochimica Acta, Vol. 43, pp.

82-83,1988, Choppin, G.n., and J N. Mathur, " Hydrolysis of actinyl(VI) cations," Radiochimica Acca, Vol.

52I53, pp. 25-28,1991.

Codell, R.B., and W M. Murphy, " Geochemical model for "C transport in unsaturated rock," in Proceedings of the Third International Conference on High-Level Radioactive Waste Management, American Nuclear Society, La Grange Park, Illinois, pp. 1,959-1,965,1992.

Couture, R.A., " Steam rapidly reduces the swelling capacity of bentonite," Nature, Vol., 318, pp.

50-52,1985.

Cox, B., " Environmentally induced cracking of zirconium alloys-A review," Joumal of Nuclear Materials, Vol.,170, pp.1-23,1990.

66

^

l Cragnolino, G.A., and J.R. Galvele, " Anodic behavior and , 'ing of zirconium and Zircaloy-4 in I aqueous solutions of sodium chloride," in R.P. Frankenthal and J. Kruger (eds.), Passiwty of Metals, The Electrochemical Society, Princeton, New Jersey, pp. 1,053-1,057,1977. l l

Cragnolino, G.A., D S. Dunn, and N. Sridhar, " Environmental factors in the stress corrosion cracking of type 316L stainless steel and alloy 825 in chloride solutions," Corrosion, Vol. 52(3),

pp.194-203,1996.

Davis, J.A., and D.B. Kent, " Surface complexation modeling in aqueous geochemistry,"in M.F.

Hochella, Jr., and A F. White (eds.), Reviews in Mineralogy: Volume 23 Mineral-Water Interface Geochemistry, Mineralogical Society of America, Washington, DC, pp. 177-260, 1990 Delany, J.M., " Reaction of Topopah Spnng Tuff with J-13 Water: A Geochemical Modeling Approach Using EO3/6 Reaction Path Code," UCRL-53631, Lawrence Livermore National Laboratory, Livermore, California,1985.

de Marsily, G., Quantitative Hydrogeology: Ground > vater Hydrology for Engineers, Academic Press, Inc , New York,1986.

Dubessy, J.. et al., "Radiolysis evidenced by H2-0 2-bearing fluid inclusions in three uranium deposits," Geochimica et Cosmochimica Acta, Vol 52, pp.1,155-1,167,1988.

Ebert, W.L., and J.K. Bates, "A companson of g! ass reaction at high and low SA/V: PCT vs MCC-1," in Proceedings of Third International Conference on High-Level Radioactive Waste Management, American Nuclear Society, La Grange Park, Illinois, Vol.1, pp. 934-942,1992.

Fehiner, F.P., Low-Temperature Oxidation: The Role of Vitreous Oxides, John Wiley & Sons, New York,1986.

Finn, P.A., et al., " Colloidal products and actinide species in teachate from spent nuclear fuel,"

Radiochimica Acta, Vol. 66/67, pp.189-195,1994a.

Finn, P.A., et al., "The effect of fuel type in saturated spent fuel tests," in Proceedings of the Fifth Internat 3nal Co-forence on High LevelIndioactive Waste Management. American Nuclear Soc,ety, La brange Park, Illinois, Vct _, pp. 1,080-1,087,1994b.

Ford, F.P., " Stress corrosion cracking of iron-base alloys in aqueous environments," in C.L. Briant and S.K. Banerji (eds.), Treatise on Materials Science and Technology: Volume 25 Embrittlement of Engineering Materials, Academic Press, New York, pp. 235-274,1983.

Glasser, F P., et al., "The chemical environment in cements," in C.M, Jantzen, J.A. Stone, and R.C. Ewing (eds.), Scientific Basis for Nuclear Waste Management Vill, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 44, pp. 849-858,1985.

Glasser, F.P., D.E. Macphee, and E.E. Lachowski, " Solubility modeling of cements: Implications for radioactive waste immobilization," in J.K. Bates and W.B. Seefeldt (eds.) Scientific Basis for 67

l'* .

l Nuclear Waste Management X, Materials Research Society, Pittsburgh, Pennsylvania, l Symposium Proceedings 84, pp 331-341,1987.

l l Goldberg, S., H.S. Forster, and E.L. Heick, " Flocculation of illite / kaolinite and illite / montmorillonite mixtures as affected by sodium adsorption ratio and pH," Clays and Clay Minerals, Vol. 39, pp. 375-380,1991.

I Grambow, B., " Spent Fuel Dissolution and Oxidation: An Evaluation of Literature Data,"

SKB Technical Report 89-13, Swedish Nuclear Fuel and Waste Management Co., Stockholm, Sweden,1989.

Gray, W.J., H.R. Leider, and S.A. Steward " Parametric study of LWR spent fuel dissolution kinetics," Journal of Nuclear Materials, Vol.190, pp. 46-52,1992.

Heimann, R.B., " Interaction of cement and radioactive waste forms in multicomoonad systems tests at 200 C. Part 2: Mineralogical changes of cernent," Cement Concrete Research, Vol.

18,pp.554-560,1988.

Herzig, J.P., D.M. Leclerc, and P. Le Goff, " Flow of suspensions through porLJs media

-Application to deep filtration." Industrial and Chemical Engineering, Vol. 62 pp. 8-35,1970.

Holland H D., and L.H. Brush, " Uranium oxides in ores and spent fuels," in Proceedings of the Conference on High-Level Radioactive Solid Waste Forms, NUREGICP 005, U.S. Nuclear Regulatory Commission, Washington. DC, pp. 597-615 1980.

Hunter, R.J., Foundations of Colloid Science Volume 1, Oxford University Press, New York, 1987 Jantzen, C.M., " Nuclear waste glass durability: 1. Predicting environmental response from thermodynamic (Pourbaix) diagrams," Journal of American Ceramic Society, Vol. 75, pp.

2,433-2,448,1992.

l' Johnson, G.K., et al., " Thermodynamic studies of zeolites: Clinoptilolite," Journal of Chemical Thermodynamics, Vol. 23, pp. 475-484,1991.

Kallay, N., E. Barouch and E. Matijevic "Diffu:.:onal detachment of colloidal particles from solid / solution interfaces," Advances in Colloids and Interface Science, Vol. 27, pp.1-42,1987.

Kerrisk, J.F., " Reaction-Path Calculations of Groundwater Chemistry and Mineral Formation at Rainier Mesa, Nevada," LA-10560-MS, Los Alamos National Laboratory, Los Alamos, New Mexico,1983.

Kerrisk, J.F., " Groundwater Chemistry at Yucca Mountain, Nevada, and Vicinity," LA-10-29-MS, Los Alamos National Laboratory, Los Alamos, New Mexico,1987.

l Kessler, J. and R. McGuire, " Yucca Mountain Total System Performance Assessment, Phase 3," EPRI TR-107191, Electric Power Research Institute, Palo Alto, California,1996.

68 sl l

l l

I

l Knauss, K.,"Zeolitization of glassy Topopah Spring tuff under hydrothermal conditions,"in J:K. Bates and W.B. Seefeldt (eds.), Scientific Basis for Nuclear Waste Management X, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 84, pp.

737-745,1987.

I Knauss, K., et al., " Hydrothermal Interaction of Topopah Spring Tuff With J-13 Water as a Function of Temperature," UCRL-90853, Lawrence Livermore National Laboratory, Livermore, California,1984.

Knauss, K.G., W.J. Beiriger, and D.W. Peifer, " Hydrothermal Interaction of Solid Wafers of Topopah Spring Tuff with J-13 Water at 90* and 150' C Using Dickson-Type, Gold-Bag Rocking Autoclaves. Long-Term Experiments," UCRL-53722, Lawrence Livermore National Laboratory, Livermore, California,1987.

Kohler, M., E. Wieland, and J.O. Leckie, " Metal-ligand-surface interactions during sorption of uranyl and neptunyl on oxides and silicates." in Y K. Kharaka and A.S. Maest (eds.),

Proceedings of the 7th International Symposium on Water-Rock Interaction-WRI-7, A.A.

Balkema, Rotterdam, Netherlands, Vol 1,51-54,1992.

Langmuir, D , " Overview of Coupled Processes with Emphasis in Geochemistry,' in C.-F. Tsang (ed ), Coupled Processes Associated With Nuclear Waste Repository. Academic Press, New York, pp.67-101,1987.

Lichtner, P.C., " Estimate of Near-Field Chemistry Based on MULTIFLO Calculations," Al 20-5708-561-793. Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas.1997 Lichtner. P.C., and J.C. Walton, "Near-Field Liquid-Vapor Transport in a Partially Saturated High-Level Nuclear Waste Repository,' CNWRA 94-022, Prepared for U.S Nuclear Regulatory Commission by the Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas,1994.

Lichtner, P.C., and J. Eikenberg, " Propagation of a Hyperalkaline Plume into the Geological Barrier Surrounding a Radioactive Waste Repository," PSI Report No. 95-01, Paul Scherrer institute, Wurenlingen, Switzerland,1995.

Lichtner, P.C., and M S. Seth. " Multiphase-multicomponent nonisothermal reactive transport in partia'i, saturated oorous media," in Proceedings of the International Conferenc a .n Deep Geologica/ Disposal of Radioactive Waste, Canadian Nuclear Society, Winnipeg, pp. 3-133 to 3-142,1996a.

Lichtner, P.C., and M.S Seth, " User's Manual for MULTIFLO: Part 11 MULTIFLO 10 and GEM 1.0, Multicomponent-multiphase reactive transport modet,"CNWRA 96-010, Prepared for U.S.

Nuclear Regulatory Commission by the Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas,1996b.

Lichtner, P.C., and D. R. Turner, " Critique of DOE's Near-Field Environment Modeling for the l

Proposed HLW Repository at Yucca Mountain, Nevada," Al 20-5708-561-791, Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas,1997, 69

-_o

s ,

Lichtner, P.C., R.T. Pabalan, C.I. Steefei, "Prelimin " Evaluation of the Effects of Alkaline Plume Migration on the Near-Field Environment of a High-Level Waste Geologic Repository,"

Prepared for U.S. Nuclear Regulatory Commission by the Center for Nuclear Waste Regulatory Analyses, San Antonio, Texan,1997.

Lieser, K. H., and C. Bauscher, " Technetium in the hydrosphere and in the geosphere. II.

Influence of pH, of complexing agents and of some minerals on the sorption of technetium,"

Radiochimica Acta, Vol. 44/45, pp.125-128,1988.

Light, W.B., et al., " Analytical models for C-14 transport in a partially saturated, fractured, porous media," in Proceedings of Nuclear Waste Isolation in the Unsaturated Zone: Focus '89, American Nuclear Society, La Grange Park, Illinois, pp. 271-277,1989.

Lin, W., and W.D. Daily, " Hydrological properties of Topopah Spring tuff under a thermal gradient: Laboratory results," International Journal of Rock Mechanics and Mineral Science and Geomechanical Abstrats, Vol. 27, pp. 373-385,1990.

Machesky, M.L., D.A Palmer, and D.J. Wc Wows <i, " Hydrogen ion adsorption at the rutile-water interface to 25v C," Geochimica et Cosmochimica Acta Vol. 58, pp. 5,627-5,632,1994 Maiti, T.C.. M.R Smith, and J.C. Laul. " Colloid formation study of U, Th. Ra, Pb, Po, Sr. Rb, and Cs in bnny (high ionic strength) groundwater: Analog study for waste disposalf Nuclear Technology. Vol. 84, pp. 82-87,1989.

Manteufel, R.D. et al., "A Literature Review of Coupled Thermal-Hydrologic-Mechanical-Chemical Processes Pertinent to the Proposed Repository at Yucca Mountain," NUREG/CR-6021, U.S. Nuclear Regulatory Commission, Washington, DC,1993.

McCarthy, J F., and J.M. Zachara. " Subsurface transport of contaminants," Environmental Sciences Technology, Vol. 23, pp. 496-502,1989.

McCauley, R.A., Corrosion of Ceramics, Marcel Dekker, New York,1995.

McDowell-Boyer, L M., J.R. Hunt. and N. Sitar, " Particle transport through porous media,"

Water Resources Research, Vol. 22, pp.1,901-1921,1986.

Meijer, A.," Yucca Mountain Project Far-Field Sorption Studies and Data Needs," LA-11671-MS, Los Alamos National Laboratory, Los Alamos, New Mexico,1990.

Murphy, W.M., " Geochemical models for gas-water-rock interactions in a proposed nuclear waste repository at Yucca Mountain, Nevada," in Proceedings of Site Characterization and Model Validation: Focus '93, American Nuclear Society, La Grange Park, Illinois, pp. 115-121, 1993.

Murphy, W.M., " Contributions of thermodynamic and mass transport modeling to evaluation of groundwater flow and groundwater travel time at Yucca Mountain, Nevada," in T. Murakami and 70 L_______________ __

r R.C. Ewing (eds.), Scientific Basis for Nuclear Waste Mananment XVill, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 353, pp. 419-426,1995.

Murphy, W.M., " Retrograde solubilities of source term phases," in W.J. Gray and I.R. Triay (eds.), Scientific Basis for Nuclear Waste Management XX, Materials Research Society, '

l Pittsburgh, Pennsylvania, Symposium Proceedings 465, pp. 713-720,1997.

l Murphy, W.M., and E.C. Pearcy, " Source-term constraints for the proposed repository at Yucca Mountain, Nevada, derived from the natural analog at Pe6a Blanca, Mexico," in C.G. Sombret

' (ed.), Scientific Basis for Nuclear Waste Management XV, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 257, pp. 521-527,1992.

l I

Murphy, W.M., and R.T. Pabalan, "GeochemicalInvestigations Related to the Yucca Mountain Environment and Potential Nuclear Waste Repository," NUREG/CR-6288, U.S. Nuclear Regulatory Commission, Washirrton, DC,1994.

l Murphy, W.M., and J.D. Prikryl, "High-Level waste near-field processes and variations " in B.

Sagar (ed.), Report on Research Activities July-December 1995, CNWRA 95-02S, Prepar:d for U.S. Nuclear Regulatory by the Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas, pp.10-1 to 10-14,1996.

Murphy, W.M., et al , " Reaction kinetics and thermodynamics of aqueous dissolution and growth of analcime and Na-Clinoptilolite at 25 'C," American Journalof Science, Vol. 296, pp.

128-186,1996.

Murphy, W M., E.C. Pearcy, and D.A. Pickett, " Natural analog studies at Pe6a Blanca and Santorini." Seventh CEC Natural Analogue Working Group Meeting Proceedings, Commission of the European Communities, Luxembourg, in press,1997.

Nash, J.T., H.C. Granger, and S.S. Adams, " Geology and concepts of genesis of important types of uranium deposits," in B.J. Skinner (ed.), Economic Geology 75th Anniversary Volume 1905-1980,pp.63-116,1981.

National Research Council, Technical Bases fo' Yucca Mountain Standards, National Academy Press, Was*gton, DC,1995.

l Neall, F.B., "Modeling the long-term chemical e.alution of cement-groundwater systems," in W.M. Murphy and D.A. Knecht (eds.), Scientific Basis for Nuclear Waste Management XIX, Materials Research Society, Pittsburgh, Pennsylvania. Symposium Proceedings 412, pp.

483-490,1996.

Needes, C.R.S., M.J. Nicol, and N.P. Finkelstein, " Electrochemical model for the teaching of uranium dioxide: 2-alkaline carbonate media," in A.R. Burkin (ed.) Leaching and Reduction in Hydrometallurgy, The Institution of Mining and Metallurgy, London, United Kingdom, pp.12-19, 1975.

71

I Nishikata, A., et al., " Monitoring of corrosion rate of carbon steel under wet / dry cycle conditions and its corrosion mechanism,". Corrosion Engineering. Vol. 43, pp. 225-234,1994.

Nitao, J.J.," Numerical Modeling of the Thermal and Hydrological Environment Around a Nuclear Waste Package Using the Equivalent Continuum Approximation: Horizontal Emplacement," UCID-21444, Lawrence Livermore National Laboratory, Livermore, California, l

1988.

Ogard, A E., and J.F. Kerrisk, " Groundwater Chemistry Along Flow Paths Between A Proposed l

Repository Site and the Accessible Environment," LA-10188-MS, Los Alamos National l Laboratory. Los Alamos, New Mexico,1984.

Olofsson, U., et al.,' Formation and properties of americium colioids in aqueous systems,"in S.V. Topp (ed.), Scientific Basis for Nuclear Waste Management IV. Materials Research Society, Pittsburgh, Pennsylvania. Symposium Proceedings 6, pp.191-198,1982a.

Olofsson, U., et al., " Properties and mobilities of actinide colloids in geologic systems." in W.

Lutze (ed.), Scientific Basis for Radioactive Waste Management V. Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 11, pp. 755-764,1982b.

Organ. R.M. "The current status of the treatment of corroded metal artifacts,"in B.F. Brown (ed.), Corrosior, and Metal Artifacts NBS Special Publication 479, U.S. Government Printing Office. Washington, DC,1977.

Pabalan. R.T , and D R. Turner ' Sorption Modeling for HLW Performance Assessment 'in B.

Sagar (ed.), NRC High-Level Radioactive Waste Research at CNWRA January Through June

{

1993, CNWRA 92-02S, Prepared for the U.S. Nuclear Regulatory Commission by the Center for Nuclear Waste Regulatory Analyses. San Antonio, Texas, pp. 8-1 to 8-18,1993.

Pabalan. R T., and D R Turner, " Uranium (6+) sorption on montmorillonite: Experimental and surface complexation modeling study," Aqueous Geochemistry, Vol. 2, pp. 203-226,1997.

Pabalan, R.T., et al., " Experimental study of uranium (6+) sorption on the zeolite mineral clinoptitolite," in C. Interrante and R. Pabala (eds.), Scientific Basis for Nuclear Waste Management XV/, Materials Re:earch Society Pittsbumh, Pennsylvania, Symnosium Proceedings 294, pp. 777 ~.~ 2,1993.

l l

Paquette, J., and R.J. Lemire. "A description of the chemistry of aqueous solutions of uranium l and plutonium to 200 *C using potential-pH diagrams," Nuclear Science and Technology, Vol.

79,pp.26-48,1981.

Pearcy, E. C., et al., " Alteration of uraninite from the Nopal I deposit, Pe6a Blanca District, Chihuahua, Mexico, compared to degradation of spent nuclear fuelin the proposed US high-level nuclear waste repository at Yucca Mountain, Nevada " Applied Geochemistry, Vol. 9, pp.

713-732,1994.

72 i

, 1 1

Pearcy, E.C., J.D. Prikryl, and B.W. Leslie, " Uranium transport through fractured silicic tuff and relative retention in areas with distinct fracture characteristics," Applied Geochemistry, Vol.10, pp.685-704,1995.

t Pierce, A.P., J.W. Mytton, and G.B. Gott, " Radioactive elements and their daughter products in I the Texas Panhandle and other oil and gas fields in the United States", in Geology of Uranium and Thorium, International Conference,1955, U.S. Geological Survey (USGS), Washington,

{1 DC, USGS Professional Paper 300, pp. 527-532,1955.

Pierce, A.P., G.B. Gott, and J.W. Mytton, " Uranium and helium in the Panhandle gas field, j Texas, and adjacent areas," U.S. Geological Survey, Washington, DC, USGS Professional  !

Paper 454-G, pp. G1-57,1964. {

l Pruess, K., J.S Y. Wang, and Y.W. Tsang, "On thermohydrologic conditions near high-level nuclear wastes emplaced in partially saturated fractured tuff.1. Simulation studies with explicit consideration of fracture effec %," Water Resources Research, Vol. 26, pp.1,23.5-1,248,10:0 {

i Ransom, B., ana H.C. Helgeson, " Estimation of the standard molal heat capacities, entropies, and volumes of 2:1 clay minerals," Geochimica et Cosmochimica Acta, Vol. 58, pp.

4,537-4,547,1994.

Reardon, E.J.," Problems and approaches to the prediction of the chemical composition in cement / water systems," Waste Management, Vol.12, pp 221-239,1992.

(

Reed, D.T., and R A. Van Konynenburg, " Radiation Effects", :n D G. Wilder (ed ), Preliminary Near-Field Environment Report Volume ll: Scientific Overview of Near-Field Environment and Phenomena, UCRL-LR-107476 Lawrence Livermore National Laboratory, Livermore, California, pp. 73-80,1993.

Rimstidt, J.D., W.D. Newcomb, and D.L. Shettel. Jr., "A vertical thermal gradient experiment to simulate conditions in vapor dominated geothermal systems, epithermal gold deposits, and high-level radioactive repositories in unsaturated media,"in Proceedings of the 4th Intemational Symposium on Water-Rock Interaction. A.A. Balkema, Rotterdam, Netherlands, pp. 585-588, 1989 Robert, M., and M. Terce, "E'fect of gels and coaungs on clay mineral chemical propeuies," in B. Bar-Yosef, N.J. Barrow, and J. Goldshmid (eds.), Inorganic Contaminants in the Vadose Zone: Ecological Studies Volume 74, Springer-Verlag, New York, pp. 57-71,1989.

Ryan, J.N., and P.M. Gschwend " Colloid mobilization in two Atlantic coastal plain aquifers:

Field studies," Water Resources Research, Vol. 26, pp. 307-322,1990.

Santanan, L., S. Raghavan, and B.A. Chin, "Zircaloy cladding rupture during rer ository storage," Nuclear Technology, Vol. 97, pp. 316-322,1992.

73

Savage, D.," Review of Potential Effects of Alkalin, olume Migration from a Cementitious Repository for Radioactive Waste," R&D Technical Report P60, Environmental Agency, Bristol, United Kingdom,1997.

Sherard, J.L., L.P. Dunnigan, and J. R. Talbot, " Basic properties of sand and gravel filters."

Journal of Geotechnical Engineering, Vol.110, pp. 684-700,1984.

Shoesmith, D.W., et al., "The corrosion of nuclear fuel (UO 2) in oxygenated solutions,"

Corrosion Science, Vol. 29, pp.1,115-1,128,1989.

Spinks, J.W.T . and R.J. Woods, introduction to Radiation Chemistry, Wiley-interscience, New York,1976.

Sridhar N., and D.S. Dunn, "Effect of applied potential on changes in solution chemistry inside crevices on type 304L stainless steel and alloy 825 " The Journal of Science and Engineering Corrosion, Vol. 50(11), pp. 357-272,1994.

Sridhar N., et al.," Review of Degradation Modes of Alternate Container Designs and Materials," CNWRA x-010. San Antonio, TX: Center for Nuclear Waste Regulatory Analyses.

1994 Sridhar, N., P.C. Lichtner. and D. Dunn, 'The Evolution of Chemistry inside Disbonded Coatings of Catholically Protected Steel Pipeline - Moceling and Experimental Studies " Final Report of Internal Research Project 20-9932. Southwest Research Institute, San Antonio, Texas,1996.

Steefel, C., and P.C. Lichtner, " Diffusion and reaction in rock matrix bordering a hyperalkaline fluid-filled fracture " Geochimica et Cosmochimica Acta, Vol. 58, pp. 3,595-3,612,1994.

Suarez, D.L., et al., 'Effect of pH on saturated hydraulic conductivity and soil dispersion," Soil Science Society of America Journal, Vol. 48, pp. 50-55,1984.

Sunder, S., and H. Christensen, " Gamma radiolysis of water solutions relevant to the nuclear fuel waste management program," Nuclear Technology, Vol.104. pp. 403-417,1993. ,)

Sunder, S., m al., Oxidation cf UO 2fuel by the products of gamma radiolysis o, water," Journct of Nuclear Materials, Vol.190, pp. 78-86,1992.

Thomas, D.M., and I. Gudmundsson, " Advances in the study of sohds deposition in geothermal systems," Geothermics, Vol.18, pp. 5-15,1989.

Thorstenson, D.C., et al., " Physical and chemical characteristics of topographically affected airflow in an open borehole at Yucca Mountain, Nevada," in Proceedings of Nuclear Waste Isolation in the Unsaturated Zone Focus: '89, American Nuclear Soctety, La Grange Park, Illinois, pp. 256-270,1990. 1 74 I

(*.

i f* Travis, B.J., and H E. Nuttall, " Analysis of colloid transport. M L.O. Werme (ed.), Scientific Basis for Nuclear Waste Management /X, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 50, pp. 737-745,1985.

t Travis, B.J., and H.E. Nuttall, "Two-Dimensional Numerical Simulation of Geochemical

(

Transport in Yucca Mountain," LA-10532-MS,. Los Alamos National Laboratory, Los Alamos, New Mexico,1987. ,

TRW Environmental Safety Systems, Inc., " Total System Performance Assessment-1995: An Evaluation of the Potential Yucca Mountain Repository," B00000000-01717-2200-00136, Prepared for U.S. Department of Energy by the TRW Environmental Safety Systems, Inc.,

Las Vegas, Nevada,1995.

TRW Environmental Safety Systems, Inc., " Status / Summary Report for Fiscal Year 1996 Activities Within the Performance Assessment Overview Study on the Consequences of Cementitious Materials," LV.PA.COS.09/96-038, Prepared for U.S. Department d Energy by the TRW EnvironmenMI Safety Systems, Inc , Las Vegas, Nevada,1996a.

. TRW Environmental Safety Systems, Inc., " Mined Geological Disposal System Advanced l Conceptual Design Report. Vol.1-11," B00000000-01717-5705-00027, Rev. 00, Prepared for U.S. Department of Enecgy by the TRW Environmental Safety Systems, Inc., Las Vegas, Nevada,1996b TRW Environmental Safety Systems, Inc., ' Reference Design Desenption for a Geologic Repository," B00000000-01717-5707-00002, Rev. 01. Prepared for U S. Department of Energy by the TRW Environmental Safety Systems, Inc. Las Vegas, Nevada,1997.

Tsang, Y.W., and K. Pruess, "A study of thermally induced convection near a high-level nuclear waste repository in partially saturated fractured tuff," Water Resources Research, Vol. 23, pp.

1,958-1,966,1987.

Turgoose, S., "The nature of surviving iron objects. Conservation of Iron," National Maritime Museum Monograph. No. 53, National Maritime Museum, Greenwich, United Kingdom, pp.1-7, 1982.

Turgoose, S., "Strucwre, composition, and de. .noratiu, of unearthed iron objec,ts." in Uurrent Problems in the Conservation of Metal Antiquities, Tokyo National Research Institute of Cultural Properties, Tokyo, Japan, pp. 35-53,1989.

U.S. Department of Energy," Site Characterization Plan, Yucca Mountain Site, Nevada Research and Developrnent Area, NV," DOE /RW-0199, Office of Civilian Radioactive Waste Management, U.S. Department of Energy, Washington, DC,1988. .

U.S. Department of Energy, "[ Draft) Highlights of the U.S. Department of Energy's Updated l Waste Containment and Isolation Strategy for the Yucca Mountain Site," enclosure to a letter -

dated July 19,1996, from S.J. Brocoum, U.S. Department of Energy, to M.V. Federline, U.S.

Nuclear Regulatory Commission,1996a. l l

75

_.__m__-____._ ___._..__.-.__m _..__~ - . - - - - .

1 e

e U.S. Department of Energy," Civilian Radioactive Waste Management Program Plan, Revision 1," DOE /RW-0458, Revision 1, Office of Civilian Radioactive Waste Management, U.S.

Department of Energy, Washington, DC,1996b.

U.S Department of Energy, " Site Characterization Progress Report: Yucca Mountain, Nevada, April 1,1996-September 30,1996, Number 15," DOE /RW-0498 Office of Civilian Radioactive j Waste Management. U.S. Department of Energy, Washington, DC,1997a.

U.S. Department of Energy (DOE), " Responses to U.S. Nuclear Regulatory Commission (NRC)

Comments on DOE's Thermohydrology Testing and Modeling Program," letter dated July 14, 1996, from S.J. Brocoum, U.S. Department of Energy, to M..?. Bell, U.S. Nuclear Regulatory Commission, Washington, DC,1997b.

U.S. Nuclear Regulatory Commission," Staff Analysis of Public Comments on Proposed Rule 10 CFR Part 60 Disposal of High-Level Radioactive Wastes in Geologic Repositories." NUPEG-l 0804, U.S. Nuclear Regulatory Commission, Washington, DC,1983.

U.S. Nuclear Regulatory Commission, "NRC Staff Site Characterization Analysis of the Department of Energy's Site Characterization Plan, Yucca Mountain Site, Ne cada," NUREG-1347, U.S Nuclear Regulatory Commission Washington, DC,1989.

U. S. Nuclear Regulatory Commission,"NRC Staff Evaluation of DOE Response to NRC Site Characterization Analysis of the DOE Site Characterization Plan, Yucca Mountain, Nevada,"

letter dated July 31,1991, from R. Bernero, U. S. Nuclear Regulatory Commission, to R. J.

Bartlett U S. Department of Energy. Washington. DC,1991. )

U.S. Nuclear Regulatory Commission, "NRC High-Level Radioactive Waste Program Annual Progress Report: Fiscal Year 1996," NUREG/CR-6513. No.1, U.S. Nuclear Regulatory Commission, Washington, DC,1996a.

U.S. Nuclear Regulatory Commission," Transmittal of the Center for Nuclear Waste Regulatory Analyces Detailed Report Related to the Audit Review of the U.S. Department of Energy's 1995 Total System Performance Assessment," letter and enclosure dated November 5,1996, from J.H. Austin, U.S. Nuclear Regulatory Commission, to R A. Milner, U. S. Department of Energy, Washington, DC,1996F U.S. Nuclear Regulatory Commission, " Branch Technical Position on the Use of Expert Elicitation in the High-Level Radioactive Waste Program," NUREG-1563, U.S. Nuclear Regulatory Commission, Washington, DC,1996c.

U.S. Nuclear Regulatory Commission," Comments on the Department of Energy Thermohydrology Testing and Modeling Program," letter dated January 23,1997, from M.J.

Bell, U.S. Nuclear Regulatory Commission, to S.J. Brocoum, U.S. Department of Energy, Las Vegas, Nevada,1997.

van Cappellen, P., et al., "A surface complexation model of the carbonate mineral-aqueous solution interface," Geochimica et Cosmochimica Acta, Vol. 57, pp. 3,505-3,518,1993.

76 i

)

e .

Van Konynenburg, R.A., " Radiation Chemical Effects in Experiments to Study the Reaction of Glass in an Environment of Gamma-trradiated Air, Groundwater, and Tuff," UCRL-53719.

Lawrence Livermore National Laboratory, Livermore, California,1986.

Vaughan, P.J., " Analysis of permeability reduction during flow of heated, aqueous fluid through Westerly Granite," in C.-F. Tsang (ed.), Coupled Processes Associated with Nuclear Waste Repositories, Academic Press, New York, pp. 529-539,1987.

Verma, K., and K. Pruess, "Thermohydrological conditions and silica redistribution near high-level nuclear waste emplaced in saturated geological formations " Journal of Geophysical Research, Vol. 93(B2), pp.1,159-1,173,1988.

Walton, F.B., J.P.M. Ross, and D.G. Johnke, "The effects of simultaneous heat and mass transport on radionuclides migration,"in C.M. Jantzen, J.A. Stone, and R.C Ewing (eds.),

Scientific Basis /or Nuclone waste Management V///. Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 44, pp. 663--672,1985.

Wan, J., and J.L. Wilson, " Visualization of the role of the gas-water interface on the fate and transport of colloids in porous media," water Resources Research, Vol. 30, pp.11-23,1994.

Wescott. R.G., et al., "NRC lterative Performance Assessment Phase 2," NUREC 1464, U.S j Nuclear Regulatory Commission, Washington, DC,1995.

White, A.F . H.C. Claassen, and L.V. Benson, "The effect of dissolution of vclcanic glass on the water chemistry in a tuffaceous aquifer, Rainier Mesa Nevada," Water-Supply Paper 1535 0, U.S. Geological Survey, Washington, DC,1980.

Wilder, D.G., "Near-Field Altered-Zone Environment Report Volume ll," UCRL-LR-124998, Lawrence Livermore National Laboratory, Livermore, California,1996.

Wilson, C.N., "Results from NNWSI Series 3 Spent Fuel Dissolution Tests," PNL-7170, Pacific Northwest Laboratory, Richland, Washington,1990.

Wilson, M.L., et al., " Total System Performance Assessment for Yucca Mountain - SNL Second Iteration (TSPA-93)," SANDQ3-2675, Prepared far U.S. Department of Energy by Sandia Nationa' Laboratory. Albuquerque, New Mexico,1994.

Wolery. T.J., "EQ3/6, A Software Package for Geochemical Modeling of Aqueous Systems: )

Package Overview and installation Guide (Version 7.0)," UCRL-MA-110662,  !

Lawrence Livermore National Laboratory, Livermore, California,1992.

77

9 f e Wronkiewicz, D.J., J.E. Young, and J.K. Bates, "Ef "s of alpha and gamma radiation on glass reaction in an unsaturated environment," in T. Abrajano, Jr., and L.H. Johnson (eds.), Scientific Basis for Nuclear Waste Management XIV, Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 212, pp.99-106,1991.

Wronkiewicz, D.J., et al., " Uranium release and secondary phase formation during unsaturated testing of UO, at 90 C," Joumal of Nuclear Materials, Vol.190, pp.107-126,1992.

Wronkiewicz, D.J., et al.," Effects of radiation exposure on glass alteration in a steam environment," in C.G. Interrante, and R.T. Pabalan (eds.), Scientific Basis for Nuclear Waste Management XV/. Materials Research Society, Pittsburgh, Pennsylvania, Symposium Proceedings 294, pp.183-190,1993.

Yang, l.C., " Flow and transpod through unsaturated rock-data from two test holes, Yucca Mountain, Nevada," in Proceedings of the Third Intemational Conference on High Level Radioactive Waste M9agement, American Nuclear Society, La Grange Park, liiinois, pp.

732-737,1992.

Yang, l.C., C.A. Peted, and D.C Thorstenson, " Carbon isotopic data from test hole USW UZ-1, l Yucca Mountain, Nevada," in Proceedings of the Fourth Intemational Conference on High Level I Radioactive Waste Management American Nuclear Socioty, La Grange Park, Illinois, pp.

i 401-406,1993.

Yang, l.C., G.W. Rattray, and P Yu, " Interpretation of Chemical and Isotopic Data from j Boreholes in the Unsaturated Zone at Yucca Mountain Nevada," Water Resources '

Investigations Report 96-4058. U.S. Geological Survey, Denver, Colorado,1996a.

Yang, l.C., et al., "Hydrochemical investigations and Geochemical Modehng in Characterizing the Unsaturated Zone at Yucca Mountain, Nevada," U.S. Geological Survey, Denver, Colorado, 1996b.

Zachara J.M., and J.P. McKinley, " Influence of hydrolysis on the sorption of metal cations by smectites: Importance of edge coordination reactions," Aquatic Sciences, Vc!. 55, pp. 250-261, 1993.

l l

i 78 r ..

- 4 4 .

d _ _ r o /

_ _ , e l

s ) s , ty

_ _ Nr e g Rt ai n N, , s e

_ _ f wp R, f i p od r n f io l u

n u op o 3 j _ nup oo n

&g r r

_ uri o ,, n a

_ lu gi e io t

ioa l

e E ' n) di nw lu s tai c

N _

_ R i( i , t o di , ci or f

_ E i t

_ [ (l c( o _

l a i

P e u l

n S s lc io _

O o a _ ) > )

t I DC te h l u

_ B I  % _

s a o

- v

( ( _ f o r t

r E o o p

_ y s f t

n tum i s n a T

) _ b in a

b a n

n ie trh r

s a K

_ _ c ih if s .

- ol a s a o s e r o r a i t

_ t s pv tn n h n l

c: e a ei T e

_ r ie t

l _ ( \ 7 ( f ( . me m J _ t n

l e e E )e Cs

_ h J DRL_h > mye T YNoD _ E R ] _ in we s m o

s s k e

F RA _

~ E t r

- on no i n i,u s d s e A OMl T Ru a _

I I

P _.

,l]i

_ gop s

- c n tnoi i

n , i.

to i ,,,&

i a h e g t

I _ S n R S O.i O F.i d

v _

O E

Z S gT a

r et ic n u ad a c d , s r

te a, a n e ci nh a g l

D PRd _

G N lor v p o

r pr

, e mi r h o e EEn ._

t

_(

f RP( I _ f _ _

r h r

. _ Y . e p

t t

r t

o

,I gop _

e s m e u t

s a l r t

, n a r u s p

'/_

~ Z g a r tn o t

c y n

_ a p a s i

~ UgT m m f r

s

_ / A _/ .

e n t

o n l

a e n &it, i t oi d

_ _ n

^

e 3 - o t v

_ J -

r u la i, or i i i

t a

r o r

cw fop t

ta ,,

D N a o pir d

r E

R d

r e J_ _ C trl s a

sao a te r mt a n E e sr i e e g m

_ l EM n e i

_ i

_ ia n

. NE ig rr Tn  ; d or G T I

n a -

y N /,_

ti s

_ S EB d. i o i) t r g i t

ni v NY / m itp,yf ,ia s i

.imtc n

s im wn ES u

_/

_ u,k c tel oE f k h r c ea a s i, or r y d _

" , ht i

c n it wd

'r e 'v d

,d os eli

" u t "r l a

,n &cm r r sb a l ol e

" r' a c(

,a yef o e lu F F i

' e=

_ in g t t leo r s

_ pe apn ia e h h tnwt

" mc hwt i

s d

_ t p _ e c e

i u

a uf a qo Nn Ra 1

e " t

( -

_ s m _ [m ,, a A D)- k t e

_ nt s _ e n r e y N r u

si o- n s

_ S e e w _

MIO ig d s e ob

- _ M u n m p u _ S F E l cf e mS T ET _

T ne ar o f N TC _

L M _ S I

( DF Co _

E SA A E _ Y _ M YR T T S ST

_ S _ B _ YE BS OY ELF UB

_ U _

TS S KEOSA

_