Technical Basis Document No. 7: In-Package Environment and Waste Form Degradation and Solubility Revision 1 


July 2004 

1. INTRODUCTION 
This technical basis document summarizes the technical basis and conceptual understanding of 
the chemical environment within breached waste packages after water has come into contact 
with the waste form, the subsequent degradation of the waste forms, and the mobilization of 
radionuclides from degraded waste forms. 
This document is one in a series of technical basis documents that are being prepared for the 
different components relevant to predicting the likely postclosure performance of the Yucca 
Mountain repository system. The relationship of in-package chemistry, waste form degradation, 
and radionuclide solubility to the other repository system components is illustrated in Figure 1-1. 
To predict the chemical environment within a breached waste package, the degradation of the 
waste form, and the mobilization of radionuclides, the models discussed in this technical basis 
document rely on input from other models shown in Figure 1-1. Specifically, models for water 
seepage into the drifts and for waste package and drip shield corrosion are used to predict the 
evolution of the chemical environment within breached waste packages, the degradation of the 
waste forms, and the mobilization of radionuclides from degraded waste forms. Output from the 
models discussed in this document provides the source term for modeling radionuclide transport 
1-1 July 2004 No. 7: In-Package Environment Revision 1 
as dissolved species and as associated with colloids through the different components of the 
engineered barrier system (EBS) (waste package, corrosion products, and invert) and subsequent 
migration through the unsaturated zone, the saturated zone, and discharge to the biosphere. The 
technical understanding of the waste package environment, waste form degradation, and 
radionuclide mobilization is based on 13 reports written to support the license application (LA): 
• Radionuclide Screening (BSC 2002a) 
• Clad Degradation — FEPs Screening Arguments (BSC 2004a) 
• Pitting Model for Zirconium-Alloyed Cladding at YMP (BSC 2003a) 
• Clad Degradation — Summary and Abstraction for LA (BSC 2003b) 
• Seismic Consequence Abstraction (BSC 2003c) 
• Initial Radionuclide Inventories (BSC 2003d) 
• CSNF Waste Form Degradation: Summary Abstraction (BSC 2003e) 
• In-Package Chemistry Abstraction (BSC 2004b) 
• DSNF and Other Waste Form Degradation Abstraction (BSC 2003f) 
• Defense HLW Glass Degradation Model (BSC 2003g) 
• Dissolved Concentration Limits of Radioactive Elements (BSC 2004c) 
• Waste Form and In-Drift Colloids-Associated Radionuclide Concentrations: Abstraction 
and Summary (BSC 2003h). 
• Igneous Intrusion Impacts on Waste Package and Waste Form (BSC 2004d). 
This technical basis document uses the information provided in these 13 reports to present a 
coherent perspective of the impact that water ingress into waste packages has on the chemical 
environment within the waste packages, the degradation of the waste forms, and the mobilization 
of radionuclides. In addition, this technical basis document provides the context for responses to 
Key Technical Issue (KTI) agreements (and associated general (GEN) items) and Additional 
Information Needed (AIN) requests made between the U.S. Nuclear Regulatory Commission 
(NRC) and the U.S. Department of Energy (DOE) related to in-package chemistry and 
commercial spent nuclear fuel (SNF) cladding. Technical responses to the following KTI 
agreements and AIN requests are provided as appendices to this document: 
• Appendix A—In-Package Chemistry Environment (Response to CLST 3.02 AIN-1, 
ENFE 3.03, TSPAI 3.14, and GEN 1.01 (Comments 116 and 126)) 
• Appendix B—Effects of Radiolysis and Engineered Materials on In-Package Chemistry 
(Response to CLST 3.03 AIN-1 and CLST 3.04 AIN-1) 
July 2004 1-2 No. 7: In-Package Environment Revision 1 
• Appendix C—Demonstration of the Adequacy of the In-Package Chemistry Model 
Results (Response to ENFE 3.04 and CLST 3.05) 
• Appendix D—Localized Corrosion and Stress Corrosion Cracking in Cladding 
(Response to CLST 3.06 AIN-1, CLST 3.07, CLST 3.08 AIN-1, CLST 3.09 AIN-1, and 
GEN 1.01 (Comment 124)) 
• Appendix E—Total System Performance Assessment Implementation of In-Package 
Chemistry (Response to TSPAI 3.08). 
1.1 UNDERSTANDING OF IN-PACKAGE CHEMISTRY, WASTE FORM 
DEGRADATION, AND RADIONUCLIDE MOBILIZATION 
As the starting point for all potential radionuclide releases to the biosphere, the source term for 
radionuclides within the EBS is an important submodel in the total system performance 
assessment (TSPA) for LA. For radionuclides to be transported through the EBS, the waste form 
must first degrade and the radionuclides mobilized. The degradation of the waste form and the 
mobilization of radionuclides within the waste package are herein jointly called the waste form 
degradation model. The function of the waste form degradation model is to estimate the 
availability and mobilization of radionuclides that are used as input by the EBS transport model. 
Although radioactive waste can degrade by oxidation without water, radionuclides cannot be 
transported in the absence of water, and water must be present for the chemistry in the package 
to be modeled. Therefore, the models presented in this document assume that water will be 
present in the waste package. In some cases (e.g., in-package chemistry), the quantity of water 
assumed to be present in the waste package is dictated by the needs of the submodel and may not 
be consistent with the quantity of water assumed to be available in other related models. 
The waste form degradation model consists of eight components, shown in Figure 1-2. Five of 
those components determine the degradation rate of waste forms: (1) in-package chemistry; 
(2) degradation of commercial SNF cladding, (3) degradation of commercial SNF, 
(4) degradation of U.S. Department of Energy SNF, and (5) degradation of high-level radioactive 
waste (HLW) glass. The other three components determine the mobilization of radionuclides 
from the waste form into the waste package corrosion products: radionuclide inventory, 
radionuclide solubility, and colloid formation and stability. 
Understanding the source term for radionuclides within the EBS begins with the model of the 
average radionuclide inventory for three generic waste forms: commercial SNF, DOE SNF, and 
HLW encapsulated in borosilicate glass. A fourth waste form, naval SNF, is represented by 
commercial SNF. The three waste forms are placed into two waste package configurations: a 
commercial SNF waste package with various numbers of assemblies of SNF, and a codisposal 
waste package with typically one canister of DOE SNF and five canisters of HLW glass. Of the 
more than 100 radionuclides that will be potentially present in the initial inventory, 
28 radionuclides (from 14 elements) are identified as being important to the estimate of expected 
dose to the reasonable maximally exposed individual (RMEI) during the 10,000-year regulatory 
period. 
July 2004 1-3 No. 7: In-Package Environment Revision 1 
Figure 1-2. Components of the Waste Form Degradation Model 
The evolution of water chemistry inside a breached waste package is controlled by the waste 
package corrosion rate, waste package type (commercial SNF versus codisposed), water inflow 
rate, cladding failure fraction, waste-form corrosion rate, and temperature. Water inflow rate 
will depend on whether seepage water drips onto the waste package. For nondripping 
conditions, water will enter a breached waste package as water vapor, whereas under dripping 
conditions, water will flow through a breached waste package. The pH of water within a waste 
package will be strongly influenced by the presence of corrosion products, primarily goethite. 
This phase will provide sufficient surface area for complexation reactions to buffer the pH below 
8.1. 
For commercial SNF, radionuclides can be released only from those fuel rods on which the 
cladding has failed. The rate of release of radionuclides from the fuel matrix is modeled as a 
function of four variables: temperature, pH, total carbonate concentration, and oxygen fugacity 
(or partial pressure). In addition, a small fraction of radionuclides resides in the grain boundaries 
of the fuel pellets and the gap between the fuel pellet and cladding of commercial SNF. 
Radionuclides (specifically 90Sr, 137Cs, 99Tc, and 129I) residing in the gap or grain boundary are 
available for dissolution as soon as water contacts the waste. The presence of a gap fraction and 
July 2004 1-4 No. 7: In-Package Environment Revision 1 
the rapid degradation of the commercial SNF fuel matrix in an oxic environment, as is expected 
to exist in the repository, implies that the release of most radionuclides from a commercial SNF 
waste package is not limited by their availability (i.e., not limited by the waste form degradation 
rate) but by their solubility. The exceptions are 90Sr, 137Cs, 99Tc, and 129I. The solubilities of 
these four radionuclides are quite high under repository conditions; hence, the degradation rate of 
the commercial SNF fuel matrix and the initial inventory of these radionuclides in the gap and 
grain boundary influences their release rate. 
Eleven DOE SNF groups have been used to categorize the several hundred distinct types of DOE 
SNF that will eventually be emplaced in the repository. A degradation rate model consisting of 
waste form degradation and radionuclide availability rapid enough to occur within the first 
TSPA-LA time step is used for all these waste groups, except naval SNF, which is assumed to 
behave as commercial SNF. 
Many well-known processes contribute to HLW glass degradation, although there is some debate 
as to whether the dissolution process at high dissolved silica (H4SiO4) levels is controlled by 
surface reactions or by the diffusion of silica through alteration phases that form on the glass as it 
dissolves, as discussed further in Section 5.3. The glass dissolution process is controlled by the 
temperature and pH of the solution in contact with the glass. Under acidic and alkaline 
conditions, glass degradation rates increase with, respectively, decreasing and increasing pH. 
The degradation rate has an Arrhenius-type dependence on temperature. 
The solubility of seven important elements (plutonium, americium, neptunium, thorium, 
uranium, protactinium, and actinium) depends on the water chemistry inside the waste package. 
The solubility exhibits a typical “U-shaped” dependence on pH with a minimum at 
approximately neutral to mildly basic pH values, which coincides with long-term in-package 
conditions. For an eighth element, radium, the solubility is defined solely as a function of pH. 
The solubilities of the remaining five important elements (i.e., carbon, cesium, strontium, iodine, 
and technetium) are high under expected repository conditions. Hence, the availability of these 
radionuclides for transport out of the waste form is determined by the fraction of cladding that 
fails, the degradation rate of the waste form, and the gap and grain boundary inventory. 
Two sources of colloids are considered in the colloid model in the waste form: natural colloids in 
seepage groundwater and colloids generated from degradation of HLW glass, which exhibit a 
clay-like behavior. In addition, a third type of colloid, iron oxyhydroxide corrosion products, is 
assumed to form in the waste package as the waste form support structures corrode. Colloids 
formed from DOE HLW glass and naturally occurring groundwater colloids are both represented 
by smectite clay colloids, and can transport reversibly sorbed radionuclides. In addition, HLW 
glass colloids can transport embedded plutonium and americium. The stability of colloid 
suspensions is a function of ionic strength and pH. Dissolved cesium, americium, plutonium, 
protactinium, and thorium are partitioned (reversibly) between the water and groundwater 
colloids. Plutonium and americium embedded in HLW glass colloids are considered to be 
irreversibly sorbed and, thus, are not partitioned between the water and the HLW glass colloids. 
Corrosion product colloids are likely to be made up of some mixture of goethite, ferrihydrite, or 
hematite. Each solid has a high affinity for many radionuclides. Colloid suspensions are stable 
only when the ionic strength is less than 0.05 mol/L, and, at certain pH values, can be unstable at 
values of ionic strength lower than 0.05 mol/L. Colloid suspensions are believed to be stable 
July 2004 1-5 No. 7: In-Package Environment Revision 1 
only for limited periods of time and under limited combinations of physical and chemical 
conditions during the 10,000-year regulatory compliance period. Moreover, in the nominal case 
scenario, when radionuclide transport within the EBS is expected to be diffusion-dominated, 
colloid-facilitated transport of radionuclides is physically impossible because the colloids will be 
too big to diffuse. 
1.2 APPLICABILITY OF WASTE FORM DEGRADATION MODEL IN THE 
SCENARIO CLASSES 
Waste form degradation is highly dependent on the amount of water entering the waste package. 
This dictates the in-package chemistry, which, in turn, influences waste form degradation rates, 
radionuclide solubility, and colloid stability. Water reaches the repository by infiltrating the 
mountain surface, percolating through the matrix and fractures above the repository, and seeping 
into the emplacement drift. Figure 1-3 shows the advective water flux pathways into and within 
the emplacement drifts, including the EBS and, hence, the potential sources of water for waste 
form degradation and radionuclide mobilization. 
It is conservatively assumed that water will be present to dissolve and transport radionuclides 
even during periods of repository conditions that would inhibit such processes (i.e., when the 
waste package and waste form temperatures are higher than the surrounding surfaces and would 
otherwise drive moisture away). Water will be present in the repository environment, and if 
radionuclides are to leave the drift in large quantities, water must interact with fuel elements. To 
establish the limits of the water-borne dose, the broad features of waste form degradation by 
water must be modeled. If water never contacts the waste form, or if water contacts it in lower 
quantities than envisioned, TSPA predictions of dose will have been overestimated and, hence, 
conservative. 
The waste form degradation model summarized in this document is intended to capture the 
expected chemical reactions induced when water enters a failed waste package either through 
advection (flow pathway 4 in Figure 1-3) or by diffusion and to estimate radionuclide release 
from waste packages into the invert either by advection (flow pathway 6 of Figure 1-3) or 
diffusion. The amount of water that enters a package is strongly dependent on the period of 
interest (i.e., whether temperatures in the repository are above the boiling point of water) and 
whether the engineered barriers have been disrupted. These two important aspects are used to 
define the three scenario classes considered in the TSPA-LA model (BSC 2002b, Section 4): 
nominal, seismic, and igneous scenario classes. A fourth scenario class, human intrusion, is a 
stylized analysis (BSC 2002b, Section 4) because of specific simplified assumptions imposed by 
the NRC in 10 CFR Part 63. 
The models presented in this document are used in the nominal and seismic scenario classes. 
The igneous scenario class has two types of disruption: igneous intrusion of a dike into an 
emplacement drift and volcanic eruption through a waste emplacement drift. For waste packages 
that are destroyed by igneous intrusion, all the models discussed in this document are applicable 
and are used in the TSPA-LA model. For both the volcanic eruption scenario and the human 
intrusion calculations, only the radionuclide inventory model (Section 2) is used because neither 
set of calculations involves transport of radionuclides in groundwater. The scenario classes that 
July 2004 1-6 No. 7: In-Package Environment Revision 1 
implement the models presented in this document and their relationship to waste package 
modeling are summarized in Table 1-1 and are discussed next. 
Figure 1-3. Potential Liquid Advective Pathways in the Engineered Barrier System 
July 2004 1-7 No. 7: In-Package Environment Revision 1 
Table 1-1. Models Implemented in the Scenario Classes Included in the Total System Performance 
Assessment for the License Application 
Colloid 
Formation 
and Stability 
X 
Scenario Class 
Nominal scenario and 
waste packages 
unaffected by disruptive 
events in other scenario 
classes 
X Waste packages affected 
by seismic scenario 
Radionuclide 
Solubility 
X 
X 
X 
Waste Form 
Degradation 
X 
X 
X 
Chemistry 
In-Package Cladding 
Failure 
X X 
X X 
X X Waste packages affected 
by igneous intrusion 
Waste packages affected 
by volcanic eruption 
Inventory 
X 
X 
X 
X 
X Waste packages affected 
by human intrusion 
X 
July 2004 
1.2.1 Degradation in the Nominal Scenario Class 
The nominal scenario represents the collection of features, events, and processes (FEPs) most 
likely to occur during the 10,000-year regulatory period, such as climate change and heating of 
the repository (BSC 2002b, Section 4.1). Based on corrosion analysis (BSC 2003i, 
Section 6.7.1), the drip shield and waste package will remain intact for the 10,000-year 
regulatory period for all but a small fraction of the waste packages. As a result, water seeping 
into the emplacement drifts of the repository (flow pathway 1 of Figure 1-3) will be diverted 
away from all but a small fraction of the packages (flow pathway 3 of Figure 1-3). Hence, 
advective radionuclide transport through the vast majority of waste packages (flow pathway 4) 
cannot occur; only diffusive radionuclide transport through the waste package can occur in the 
10,000 years after repository closure. Radionuclide transport by diffusion is possible only for 
dissolved species. In films thinner than the diameter of the colloids, transport is physically 
hindered. In thicker films, the substantially lower diffusion coefficients for colloids relative to 
dissolved species will tend to prevent appreciable colloid transport. 
For the nominal scenario, only a small fraction of the waste packages (e.g., the probability of 
three or more failed waste packages is 0.026 (BSC 2003i, Table 46)) is expected to have small 
manufacturing weld defects near the waste package lid, allowing water vapor to diffuse into the 
package, saturate the contents of the package, and dissolve radionuclides, thereby establishing a 
pathway for diffusive radionuclide transport (Figure 1-4). 
Only the small fraction of fuel rods that arrive at the repository with perforated cladding (about 
0.1%) can lead to the release of radionuclides during the 10,000-year regulatory period. 
Additional cladding failures from other mechanisms (e.g., localized corrosion, static loading of 
rock) during the regulatory period are not anticipated. The cladding that is initially perforated is 
assumed to split along the entire length of the fuel rod. Subsequent degradation of exposed fuel 
is assumed to produce a water-saturated rind through which dissolved radionuclides diffuse, 
thereby leaving the waste matrix and entering into the corrosion products in the waste package. 
1-8 No. 7: In-Package Environment Revision 1 
10,000 years. 
Figure 1-4. Degradation of Waste Form under Nominal Scenario Class 
NOTE: For the nominal scenario, waste form degradation is limited and only diffusive release through 
manufacturing defects occurs because the drip shield and waste package remain intact for the first 
July 2004 
1.2.2 Degradation in the Seismic Disruptive Scenario 
For the seismic disruptive scenario class (BSC 2002b, Section 4.3), the drip shields, waste 
packages, and cladding can be disrupted due to vibratory ground motion, fault displacement, and 
rockfall, as discussed in Seismic Consequence Abstraction (BSC 2003c, Table 31). The most 
likely waste package failure mechanism from a seismic event is accelerated stress corrosion 
cracking in areas that exceed the residual stress threshold for Alloy 22. It is postulated that a 
dense network of stress corrosion cracks would form, and the effective area of these cracks, 
which is much less than the overall affected surface area of the waste package, is the area 
through which flow and transport occur (Figure 1-5). The percentage of cladding that is 
perforated is a function of the peak ground velocity associated with the seismic event. The 
models of how the cladding splits once it is perforated and how radionuclides are transported 
through the rind are the same in the seismic scenario as they are in the nominal scenario. For 
those waste packages associated with undamaged drip shields or not encountering dripping 
water, radionuclide releases will be due to diffusion, similar to conditions in the nominal 
scenario. 
1-9 No. 7: In-Package Environment Revision 1 
Figure 1-5. Degradation of Waste Form under the Seismic Disruptive Scenario 
1.2.3 Degradation in the Igneous Intrusion Scenario 
In the igneous intrusion scenario, a basalt dike is envisioned to intersect the repository (BSC 
2002b, Section 4.2). Damage to waste packages and waste forms is conceptually divided into 
two distinct zones. In Zone 1 (where the emplacement drifts are intersected by the basaltic dike), 
the drip shields, waste packages, and commercial SNF cladding are disrupted and breached from 
the shock wave and heat. As a result, the waste form is distributed throughout the cooled magma 
but is chemically unchanged from the nominal scenario model. The latter is a reasonable 
assumption because the melting temperatures of fuel components are typically higher than the 
expected maximum temperature of the magma. Therefore, the waste degradation rates and 
dissolved concentration of radionuclides in water in the basalt would have the same dependence 
on water chemistry as does waste not affected by the intrusion (BSC 2004d, Section 6.5.1.2). In 
Zone 2, no damage is assumed, and thus, conditions are expected to be the same as in the 
nominal scenario (BSC 2004d, Section 8.1). 
Degradation and dissolution of the waste form in breached waste packages in Zone 1 and 
degradation of the waste in Zone 2 occur once the drift cools (nominally, within roughly 
100 years). In Zone 1, the drift is assumed to be completely filled with basalt. This scenario is 
most reasonably modeled as a uniform percolation of basalt-equilibrated fluid contacting bare 
fuel components (Figure 1-6) (BSC 2004d, Section 6.3). 
1-10 July 2004 No. 7: In-Package Environment Revision 1 
Figure 1-6. Degradation under the Igneous Intrusion Scenario 
1.3 NOTE REGARDING THE STATUS OF SUPPORTING TECHNICAL 
INFORMATION 
This document was prepared using the most current information available at the time of its 
development. This technical basis document and appendices providing KTI agreement and AIN 
request responses prepared using preliminary or draft information reflect the status of Yucca 
Mountain Project scientific and design bases at the time of submittal. In some cases, this 
involved the use of draft analysis model reports and other draft references whose contents may 
change with time. Information that evolves through subsequent revisions of the analysis model 
reports and other references will be reflected in the LA as the approved analyses of record at the 
time of LA submittal. Consequently, the project will not routinely update either this technical 
basis document or its KTI agreement appendices to reflect changes in the supporting references 
prior to submittal of the LA. 
July 2004 1-11 No. 7: In-Package Environment INTENTIONALLY LEFT BLANK 
1-12 No. 7: In-Package Environment 
Revision 1 
July 2004 Revision 1 
2. RADIONUCLIDE INVENTORY 
This section summarizes the various waste forms and waste package configurations planned for 
the repository, as well as the screening process used to select the radionuclides included in the 
source term and the quantities of those radionuclides. 
The sources of data are presented, as are the uncertainties, limitations, and confidence in the 
source term model. The information in this section is drawn primarily from Radionuclide 
Screening (BSC 2002a) and Initial Radionuclide Inventories (BSC 2003d). The information 
presented in this section serves as input to the in-package chemistry model (Section 3), the 
commercial SNF cladding degradation model (Section 4), and the waste form degradation 
models (Section 5), as shown in Figure 1-2. 
2.1.1 Waste Packages 
Four types of waste are to be placed in the repository: commercial SNF, naval SNF, DOE SNF 
(not including naval SNF), and high-level radioactive waste (HLW) glass. Three packaging 
schemes will be used for these waste types. One waste package type will contain only 
commercial SNF. The second will contain only naval SNF but will be modeled as a commercial 
SNF waste package. The third waste package type, designated as a codisposal waste package, 
will contain DOE SNF and HLW in the same waste package. Figure 2-1 illustrates how waste in 
these categories will be combined and placed into waste packages. 
Commercial SNF is classified into two broad categories based on the design of the reactor that 
produced the fuel: pressurized water reactor (PWR) or boiling water reactor (BWR). 
Commercial nuclear power plants use a variety of fuels and fuel configurations in their reactor 
cores to generate power. The predominant nuclear fuel is enriched uranium dioxide. Fuel pellets 
are packed into long cylindrical fuel rods that vary in size depending on reactor design, and these 
fuel rods, which are clad in Zircaloy or stainless steel, are bundled into assemblies. The number 
of fuel rods per assembly and the number of assemblies in a reactor core vary, depending on the 
core and reactor design (i.e., PWRs or BWRs). 
DOE SNF consists of more than 250 distinct types of SNF, and, much like commercial SNF, 
radionuclide inventories vary widely depending on the history of the fuel. The large quantity of 
DOE SNF is indicative of the large number and variety of different research reactors within the 
DOE complex. The DOE fuel assemblies have been categorized by the size, shape, composition, 
and condition of the assemblies, and the size and corrosion resistance of the canister into which 
they may be loaded (DOE 2003a). The DOE SNF waste forms will be packaged in stainless 
steel canisters that will be loaded into the waste packages. 
2.1 TYPICAL WASTE PACKAGES, RELEVANT PROCESSES, AND MODELING 
ASSUMPTIONS 
July 2004 2-1 No. 7: In-Package Environment Revision 1 
Source: BSC 2003d, Figure 1. 
Figure 2-1. An Overview of Various Waste Package Types Containing Different Wastes 
The HLW in storage at DOE sites is produced by treatment of waste from weapons production 
and by the reprocessing of SNF (mostly DOE SNF). The technology for immobilization of 
HLW is vitrification in a borosilicate glass. HLW glass will come from four DOE sites and will 
be delivered to the repository in either short (about 10 ft long) or long (about 15 ft long) 
pour-canisters. The Hanford Site will produce long canisters while the Savannah River Site and 
Idaho National Engineering and Environmental Laboratory will produce short canisters. 
Additionally, a small amount of HLW glass has been produced in short canisters at the West 
Valley Demonstration Project in New York State. Because the fuels reprocessed at each of these 
sites differ, the radionuclide inventory of the HLW and resultant glass product will vary among 
the sites. 
Details regarding the design and long-term performance of naval SNF are provided in a separate 
report that has limited distribution for security purposes. Naval SNF is conservatively treated as 
commercial SNF in the TSPA-LA. 
July 2004 2-2 No. 7: In-Package Environment Based on design criteria such as structural integrity, thermal performance, criticality safety, and 
shielding properties, 10 waste packages were designed to accommodate the numerous waste 
types (DOE 2002a, Section 3.1). Five waste package designs are used to dispose of the two 
types of commercial SNF (BWR and PWR), three waste package designs are used to dispose of 
high-level radioactive waste glass from the four different sources (Hanford, West Valley, 
Savannah River, and Idaho National Engineering and Environmental Laboratory) as well as 
DOE SNF, and two waste package designs are used to dispose of naval SNF. Table 2-1 lists the 
contents of each of the 10 different waste packages and Figure 2-2 illustrates these waste 
package designs. 
Table 2-1. Waste Package Designs 
Waste Package 
Design 
21-PWR Absorber 
Plate 
21-PWR Control Rod 
12-PWR Long 
44-BWR 
24-BWR 
5-DHLW/DOE SNF 
Short 
5-DHLW/DOE SNF 
Long 
Naval SNF Short 
Naval SNF Long 
Source: DOE 2002a, Tables 3-2 and 3-3. 
Capacity: 21 commercial pressurized water reactor assemblies 38%/55% 
and an absorber plate for preventing criticality 
Capacity: 21 commercial pressurized water reactor assemblies 1%/1% 
with higher reactivity, requiring additional criticality control that 
is provided by the placement of control rods in all assemblies 
Capacity: 12 commercial pressurized water reactor assemblies 2%/2% 
and an absorber plate for preventing criticality; longer than the 
fuel assemblies placed in the 21-PWR packages. Because of 
its smaller capacity, it may also be used for fuel with higher 
reactivity or thermal output. 
Capacity: 44 commercial boiling water reactor assemblies and 25%/32% 
an absorber plate for preventing criticality. 
Capacity: 24 commercial boiling water reactor assemblies with 
higher reactivity, requiring a thicker absorber plate to prevent 
criticality than that used in the 44-BWR design 
Capacity: 5 short high-level radioactive waste canisters and 1 
short DOE SNF canister. 
Capacity: 5 long high-level radioactive waste canisters and 1 
long DOE SNF canister 
2-MCO/2-DHLW Long Capacity: 2 DOE multicanister overpacks and 2 long high-level 1%/<1% 
radioactive waste canisters. 
Capacity: 1 short naval SNF canister 
Capacity: 1 long naval SNF canister 
Description 
Approximate 
Percentage of Waste 
Packages/Approximate 
Percentage of MTHM 
1%/<1% 
14%/3% 
15%/4% 
2%/<1% 
1%/<1% 
For purposes of modeling in the TSPA-LA, two representative waste package types are 
considered: (1) a representative commercial SNF waste package and (2) a representative 
codisposal waste package designed to hold both DOE SNF and HLW glass. As shown in 
Table 2-1, the number of commercial SNF configurations dominates the number of codisposal 
waste package configurations. Both waste package designs consist of a waste package with an 
outer corrosion resistant layer of Alloy 22 and an inner layer of stainless steel for structural 
strength (BSC 2003i, Section 1). 
No. 7: In-Package Environment 
Revision 1 
July 2004 2-3 Source: DOE 2002a, Figure 3-5. 
Figure 2-2. Waste Package Configurations for Commercial Spent Nuclear Fuel, U.S. Department of 
Energy Spent Nuclear Fuel, N Reactor Fuel, and High-Level Radioactive Waste Glass 
No. 7: In-Package Environment 
Revision 1 
July 2004 2-4 Revision 1 
2.1.2 Radionuclide Screening 
Determining the initial radionuclide inventory is the foundation for estimating the radionuclide 
source term for all transport processes in the TSPA-LA model. This determination involves two 
steps. First, a screening process is used to determine those radionuclides with the highest 
contribution to expected dose.1 Those radionuclides shown not to contribute significantly to 
expected dose can be eliminated from further consideration; in practice, this involved separating 
the radionuclides that make up roughly 95% of the dose (see below) from the remainder. 
Second, the initial inventory of radionuclides is estimated for each waste form type and each 
waste package type. 
Radionuclides contained in the waste packages include fission products from reactor operations, 
actinides from neutron capture in uranium and plutonium, and activation products from neutron 
irradiation of structural materials and trace elements. Altogether, these fission products, 
actinides, and activation products include more than 100 radionuclides that may be collectively 
present in the waste packages at the time of repository closure. Many of the radionuclides are 
present in small quantities or have intrinsic physical and chemical properties that limit the 
quantities that can reach the RMEI (e.g., short half-life, low solubility, or strongly sorbing 
characteristics). Such nuclides will not be significant contributors to expected dose and will not 
pose a radiological risk to a RMEI at the point of compliance. The remaining nuclides represent 
only a small set of the total and need to be considered in the evaluation of repository postclosure 
performance. 
Radionuclides with half-lives less than 10 years that are not decay products of other 
radionuclides in the waste inventory will not contribute significantly to the dose in the 
groundwater scenarios (BSC 2002a, Assumption 5.9). As an additional constraint, radionuclides 
that are most important to expected dose were determined using the screening process of the 
National Council on Radiation Protection and Measurement. This screening was based on dose 
calculations, which in turn consider consumption of locally produced vegetables, fish, meat, and 
milk; water consumption; inadvertent ingestion of soil; gardening and shoreline activities; 
inhalation; and exposure to contaminated ground. Two screening factors were calculated for 
each radionuclide, one for scenario classes involving groundwater transport (nominal, seismic, 
and igneous intrusion), and one for the volcanic eruption scenario, which does not involve 
groundwater transport. Screening factors for groundwater transport scenario classes accounted 
for radionuclide sorption and solubility characteristics. The screening factors are described in 
Screening Models for Releases of Radionuclides to Atmosphere, Surface Water, and Ground 
(NCRP 1996) and were adjusted to reflect the local biosphere, as described in Attachments I and 
II of Radionuclide Screening (BSC 2002a). 
Radionuclides were screened by calculating a radionuclide-screening product for each 
radionuclide. The radionuclide-screening product, which is meant to be roughly proportional to 
dose, is obtained by multiplying the screening factor for each radionuclide by the activity of that 
radionuclide in the inventory. These screening products were ranked from largest to smallest, 
1 Some radionuclides that are not directly important to expected dose are still included because (1) regulatory 
requirements mandate their inclusion (40 CFR 197.30, 10 CFR 63.331), or (2) they are parents to radionuclides 
that are important to dose. 
July 2004 2-5 No. 7: In-Package Environment Revision 1 
and were then summed, starting with the largest and adding the next largest until all of the 
screening products of each contributing radionuclide were included in the sum. Radionuclides 
were then screened based on their contribution to the summed radionuclide-screening product. 
The screening was at the 95% level, meaning that ranked radionuclides that contributed up to 
95% of the summed radionuclide-screening product were considered potentially important and 
were retained for analysis (BSC 2002a, Section 5.5). 
The relative importance of individual radionuclides to expected dose was evaluated for several 
waste types, time frames, and release scenarios. Those scenarios that require groundwater 
transport (e.g., nominal, seismic, and igneous intrusion) were evaluated together as “groundwater 
scenarios.” The eruptive igneous scenario, which does not involve groundwater transport, was 
analyzed separately. Because radionuclide transport mechanisms differ between the scenarios 
that involve groundwater transport and the eruptive igneous scenario, the set of radionuclides 
identified as being important also differs between the groundwater-transport and nogroundwater-
transport scenarios. The effects of inventory abundance, radionuclide longevity, 
element solubility, and element transport affinity were considered. To evaluate the effects of 
inventory abundance, eight waste types were examined (i.e., BWR, PWR, HLW, and DOE SNF 
for both average and bounding waste forms of each type). For the 10,000-year regulatory period, 
the screening times were at 100, 200, 300, 500, 1,000, 2,000, 5,000, and 10,000 years after 
emplacement. These sets of screening times capture the main features of the changing dominant 
radionuclides and their relative activities (BSC 2002a, Section 6.2.3). To evaluate the effects of 
element solubility and transport affinity, the elements were evaluated in three solubility groups 
(highly soluble, moderately soluble, and slightly soluble to nonsoluble) and in three transport 
affinity groups (highly sorbing, moderately sorbing, and slightly sorbing to nonsorbing). The 
isotopes in each group were compared to one another for relative importance. 
The results of the screening process at the 95% level are shown in Table 2-2. The table also 
includes eight radionuclides (241Pu, 245Cm, 228Ra, 235U, 230Th, 232Th, 242Pu, and 236U) that were 
not important to expected dose but that must be included in the inventory because of regulatory 
requirements or because they are parents to radionuclides that are important to expected dose 
(BSC 2003d, Section 6.1). The rationale for inclusion of these eight radionuclides is as follows: 
• The U.S. Environmental Protection Agency and NRC regulations require consideration 
of the combined activity of 226Ra and 228Ra in groundwater (40 CFR 197.30, 
10 CFR 63.331). Therefore, 228Ra must be added to the list of radionuclides. Because 
228Ra is produced by the decay of 236U and 232Th, both 236U and 232Th must be included 
in the inventory. 230Th must be included because it decays into 226Ra. 
• 241Pu and 245Cm must be included because they decay to 241Am, which is one of the 
screened-in radionuclides. 
• 235U must be included because it decays (via short-lived 231Th) to 231Pa, which is one of 
the screened-in radionuclides. 
• 242Pu must be included because it decays into 238U, which is one of the screened-in 
radionuclides. 
July 2004 2-6 No. 7: In-Package Environment As shown in column 2 of Table 2-2, 28 isotopes of 14 elements must be included in the 
TSPA-LA model during the 10,000-year regulatory period for scenario classes involving 
groundwater transport. A subset of these 28 isotopes and 14 elements (i.e., 14 isotopes of eight 
elements, as shown in column 3 of Table 2-2) must be included in 10,000-year TSPA-LA model 
for volcanic eruption. Therefore, 28 isotopes of 14 elements are included in the initial inventory 
for TSPA-LA calculations. 
Table 2-2. Radionuclides Included in Inventory for the 10,000-Year Regulatory Period 
Radionuclide 
227Ac 
241Am 
243Am 
14C 
245Cm 
135Cs 
137Cs 
129I 
237Np 
231Pa 
238Pu 
239Pu 
240Pu 
241Pua 
242Pub 
226Ra 
228Rac 
90Sr 
99 
229 
230 
232
Tc 
Th 
Thd 
The 
232U 
233U 
234U 
235Uf 
236Ue 
238U 
Isotopes 
Groundwater Scenario Classes 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
28 
Volcanic Eruption 
(No Groundwater Transport) 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
14 
8 Elements 
Source: BSC 2002a, Table 10; BSC 2003d, Table 16 and Section 6.1. 
NOTE: a Included because precursor to 241Am. 
b Included because precursor to 238U. 
c Included for groundwater protection requirements. 
d Included because precursor to 226Ra. 
e Included because precursor to 228Ra. 
f Included because precursor to 231Pa. 
No. 7: In-Package Environment 
14 
Revision 1 
July 2004 2-7 Revision 1 
2.2 MODEL OF RADIONUCLIDE INVENTORY 
The nominal repository is designed to accommodate a total waste inventory of 70,000 MTHM. 
As designed, approximately 63,000 MTHM of the total inventory is commercial SNF and 
approximately 7,000 MTHM is DOE SNF and HLW. The radionuclide inventory for a 
representative commercial SNF and a representative codisposal waste package was derived by 
averaging the radionuclide inventories in each of the commercial SNF and codisposal waste 
package configurations over the number of waste packages per configuration. 
Approximately 221,000 commercial SNF assemblies will be disposed of in the five proposed 
commercial SNF waste package designs (see first five entries of Table 2-1). These 
configurations are designed to accommodate each assembly based on its criticality safety needs. 
Each assembly, depending on the reactor configuration, initial fuel enrichment, burnup, and the 
age of the waste (time in storage), will have a unique isotopic composition. The total number of 
commercial SNF waste packages is 7,472 (BSC 2003d, Table 17). 
Naval SNF is a robust fuel, more robust than commercial SNF and much more robust than DOE 
SNF. Releases from naval SNF waste packages are expected to be lower than those for 
commercial SNF waste packages (BSC 2001, Section 6.5). As such, naval fuel is not treated as 
DOE SNF in the codisposal waste package. For the TSPA-LA analysis, the 300 naval SNF 
waste packages have been treated as if they were commercial SNF waste packages with respect 
to their waste form degradation mechanisms and effects, as well as their initial radionuclide 
inventory (BSC 2003d, Section 6.2 and Attachment II). Naval SNF will be disposed of in two 
different waste package configurations, as shown in Figure 2-2. 
Approximately 16,000 HLW glass logs and 3,600 DOE SNF canisters will be disposed of in the 
codisposal waste package configurations. The total number of codisposal waste packages is 
3,412 (BSC 2003d, Table 17). 
The result of the radionuclide screening analysis and the initial radionuclide inventory analysis is 
the initial radionuclide inventory (in terms of mass) of those radionuclides determined to be 
important to expected dose for a representative commercial SNF waste package and a 
representative codisposal waste package. The nominal initial radionuclide inventory for each 
waste form is given in Table 2-3; uncertainty in the radionuclide inventory is discussed in 
Section 2.4. 
For certain radionuclides, the activity decreases with time as a result of simple radioactive decay; 
for others, the activity increases with time because of in-growth. For those radionuclides in the 
fission product category, the model abstraction calculates the variation of activity with time in 
accordance with the first order decay law. The inventory of short-lived fission products, such as 
90Sr and 137Cs, decreases substantially over the 10,000-year compliance period. In contrast, the 
inventories of long-lived fission products such as 14C, 99Tc, and 129I decrease only slightly over 
the compliance period. 
2-8 July 2004 No. 7: In-Package Environment Table 2-3. Nominal Initial Radionuclide Inventory for Each Waste Form 
Commercial SNF 
(g per waste package) 
2.50 × 10-6 
8.28 × 103 
1.26 × 103 
1.37 × 100 
1.77 × 101 
4.41 × 103 
5.97 × 103 
1.75 × 103 
4.63 × 103 
9.28 × 10-3 
1.54 × 103 
4.37 × 104 
2.08 × 104 
2.69 × 103 
5.34 × 103 
–b 
–b 
2.52 × 103 
7.64 × 103 
–b 
1.54 × 10-1 
–b 
1.03 × 10-2 
5.83 × 10-2 
1.77 × 103 
6.34 × 104 
3.89 × 104 
7.92 × 106 
Radionuclide 
227Ac 
241Am 
243Am 
14Ca 
245Cm 
135Cs 
137Cs 
129I 
237Np 
231Pa 
238Pu 
239Pu 
240Pu 
241Pu 
242Pu 
226Ra 
228Ra 
90Sr 
99Tc 
229Th 
230Th 
232Th 
232U 
233U 
234U 
235U 
236U 
238U 
Source: BSC 2003d, Section 7.1, Table 21. 
NOTE: Year of inventory projection: Commercial SNF, 2033; DOE SNF, 2030; and HLW, 2030. Total number of 
waste packages is 11,184 (7,772 commercial SNF and 3,412 codisposed). 
a 18% of 14C for commercial SNF resides in the hardware outside of the cladding. 
b For these radionuclides, the inventory is an unknown small number. 
For the radionuclides of actinide elements, the model abstraction accounts for the decay 
sequence and calculates the activity of each radionuclide as a function of decay and in-growth. 
Many of the actinide elements in Table 2-3 are daughter products. 
If the nuclear waste remains contained in the waste packages, the decay history of each 
radionuclide can be analytically calculated in accordance with its decay sequence and half-life 
(Figure 2-3). 
No. 7: In-Package Environment 
DOE SNF 
(g per waste package) 
1.2 × 10-3 
2.15 × 102 
6.63 × 100 
1.78 × 100 
9.11 × 10-2 
9.59 × 101 
9.57 × 101 
3.51 × 101 
8.02 × 101 
2.11 × 100 
1.23 × 101 
2.18 × 103 
4.28 × 102 
2.88 × 101 
2.97 × 101 
4.50 × 10-5 
1.49 × 10-5 
5.14 × 101 
1.56 × 102 
3.19 × 10-1 
1.16 × 10-1 
2.14 × 104 
1.26 × 100 
5.30 × 102 
4.66 × 102 
2.47 × 104 
1.23 × 103 
6.74 × 105 
2-9 
Revision 1 
HLW 
(g per waste package) 
–b 
2.07 × 10-4 
4.07 × 101 
6.24 × 10-1 
5.89 × 10-2 
1.38 × 102 
3.28 × 102 
7.89 × 101 
1.08 × 102 
1.66 × 100 
4.24 × 101 
6.06 × 102 
5.01 × 101 
1.32 × 100 
4.22 × 100 
2.63 × 10-5 
6.51 × 10-6 
1.89 × 102 
1.10 × 103 
3.58 × 10-3 
8.81 × 10-4 
3.23 × 104 
4.43 × 10-4 
2.11 × 101 
2.53 × 101 
1.53 × 103 
6.50 × 101 
2.57 × 105 
July 2004 July 2004 
Source: BSC 2002a, Table 2; Figure adapted from CRWMS M&O 2000a, Figure 3.5-6. 
Figure 2-3. Decay Histories for Radionuclides Modeled in Total System Performance Assessment for License Application 
2-10 No. 7: In-Package Environment Revision 1 
2.3 SOURCE OF DATA AND SUPPORTING DOCUMENTS 
A 1995 data submittal from the commercial utilities provided the basic information from which 
the TSPA for the site recommendation (TSPA-SR) analysis inventory for commercial SNF was 
developed. At that time, the utilities supplied historical information about reactor assembly 
discharges up through December 1995, and they provided forecasts for the next five assembly 
discharges from their reactors. With this information, a design basis waste stream was developed 
in 1999 Design Basis Waste Input Report for Commercial Spent Nuclear Fuel (CRWMS 
M&O 1999a), as well as forecasts for assembly discharges over the lifetime of each commercial 
power reactor. This same information is also used for the TSPA-LA analysis to arrive at the 
nominal values in Table 2-4. In addition, 2002 Waste Stream Projections Report (Williams 
2003) was used in conjunction with the 1995 data to develop the uncertainty bounds for the 
nominal commercial SNF inventory values. 2002 Waste Stream Projections Report (Williams 
2003) provides bounding waste stream information to capture probable limits of what might 
occur in the future with regard to fuel selection by utilities. 
Table 2-4. Uncertainty Multipliers for the Initial Radionuclide Inventory for Each Waste Form Type 
Parameter 
Isotopes 
Distribution 
Minimum 
Most Likely 
DOE SNF 
All except 238U 
Triangular 
0.45 
0.62 
2.9 
Commercial SNF 
All except 238U 
Uniform 
0.85 
N/A 
1.4 Maximum 
Source: BSC 2003d, Section 7.1, Table 22. 
NOTE: See Section 2.4 for a discussion on how uncertainty multipliers were developed. 
DOE SNF information was collected from Source Term Estimates for DOE Spent Nuclear Fuels 
(DOE 2003a). This report provided information for the 28 radionuclides of interest and provided 
the basis for the modification to the codisposal waste package configurations by the addition of 
the wide canisters. 
The HLW radionuclide inventory (in curies per canister for each radionuclide) has been 
evaluated for the borosilicate glass to be produced at the four sites that will be supplying HLW to 
the repository: Hanford Site (14,500 canisters), Savannah River Site (5,978 canisters), Idaho 
National Engineering and Environmental Laboratory (1,190 canisters), and West Valley 
Demonstration Project (300 canisters) for a total of approximately 22,000 HLW canisters 
(CRWMS M&O 2000b, Section 5.0). As this total exceeds the approximately 16,000 HLW 
canisters that will be disposed of in the mountain, each site’s total radionuclide inventory was 
proportionally assigned in accordance with its total production. 
High-Level Radioactive Waste 
All 
Triangular 
0.70 
1 
1.5 
July 2004 
2.4 UNCERTAINTIES, LIMITATIONS, AND MODEL CONFIDENCE 
An important source of uncertainty in the model of radionuclide inventory stems from the 
assumption that screening radionuclides at the 95% level is appropriate for TSPA-LA (see 
Section 2.1.2). To address this uncertainty, an additional screening was performed at the 99% 
level. At the 99% level, an additional 10 radionuclides were identified as important to dose in the 
2-11 No. 7: In-Package Environment Revision 1 
groundwater scenarios, and an additional nine radionuclides were identified as important to dose 
in the volcanic eruption scenario. Of the 10 marginally important radionuclides identified at the 
99% level for the groundwater scenarios, five are included in TSPA-LA because they are 
precursors to radionuclides identified as important at the 95% screening level (241Pu, 242Pu, 232Th, 
235 
237
U, and 236U); five are not included in the TSPA-LA (36Cl, 244Cm, 63Ni, 210Pb, and 126Sn). Of 
the nine marginally important radionuclides identified as important to dose in the volcanic 
eruption scenario, six are included in the TSPA-LA modeling of the volcanic eruption scenario 
( Np, 231Pa, 242Pu, 126Sn, 99Tc, and 228Th), while three are not included in TSPA-LA modeling 
of the volcanic eruption scenario (225Ac, 244Cm, and 225Ra). 
There are three sources of uncertainty in the radionuclide inventory common to all waste types. 
The first is due to the computational method and nuclear data used in predicting future 
radionuclide inventories (e.g., isotopic neutron cross section or decay half-life). The second 
source of uncertainty is the completeness of records kept for SNF burnup history and HLW batch 
compositions. The third source of uncertainty, which is the most difficult to quantify, is the 
uncertainty about future decisions that may influence the creation, packaging, or shipment of 
waste (BSC 2003d, Section 6.6). An analysis of these sources of uncertainty yielded the 
uncertainty multipliers shown in Table 2-4. 
2.4.1 Commercial Spent Nuclear Fuel 
An analysis of the potential error associated with the computational method and nuclear data, 
which also includes error in burnup history, has provided correction factors (i.e., ratios) to 
represent the minimum and maximum errors of 0.89 and 1.08 for the inventory of commercial 
SNF (BSC 2003d, Section 6.6.1). 
The uncertainty due to heterogeneity of waste in the repository average inventories was 
investigated by comparing average burnups of three 1999 arrival forecasts and four 2002 arrival 
forecasts (BSC 2003d, Section 6.6.1). The minimum and maximum ratios of the projected 
average burnups for these cases over those given in Table 2-4 are 0.95 and 1.3. When multiplied 
by the minimum and maximum correction factors of 0.89 and 1.08 for the computational 
method, a range of 0.85 to about 1.4 is obtained (BSC 2003d, Section 6.6.1). Since all the 
radionuclide inventory, except 238U, is highly burnup-dependent, they should not be sampled 
independently. Therefore, an uncertainty multiplier is chosen, which is sampled and then applied 
to all radionuclide inventories except 238U. In defining the probability distribution of this 
multiplier, the end points are known, but the shape of the uncertainty distribution is not. In the 
absence of further information, a uniform distribution is the distribution of choice because it 
equally weighs all possible values. In the TSPA-LA model, a uniform distribution from 0.85 
to 1.4 for an uncertainty multiplier is applied to the nominal values (provided in Table 2-3) for 
all radionuclides except 238U. The 238U uncertainty is very small and not modeled in TSPA-LA 
(BSC 2003d, Section 6.6.1). 
2.4.2 U.S. Department of Energy Spent Nuclear Fuel 
The fuel information currently available at DOE storage sites is often determined by the records 
requirements and the intended disposition path at the time the fuel was placed into storage. 
These requirements and disposition paths were often unique to each of the sites and evolved over 
2-12 July 2004 No. 7: In-Package Environment Revision 1 
time. As a result, the availability and completeness of the radionuclide inventories and 
associated documentation varies considerably for DOE SNF. Detailed characterization of these 
fuels is not necessary because the conservative source term estimate for these fuels is used for 
repository design, analyses, and licensing activities. 
A conservative estimate of this SNF inventory was developed for each of the DOE SNF storage 
sites. The inventory was generated using calculational techniques described in Methodologies for 
Calculating DOE Spent Nuclear Fuel Source Terms (DOE 2000), which uses relevant 
experimental data and confirmatory studies. Additional studies that demonstrate the validity of 
the model and underlying codes have been performed (DOE 2003a, p. 14). The result of this 
work is a database with over 500 DOE SNF types. Each DOE SNF type has entries that include 
the radionuclide inventory, number of assemblies, and number and type of canisters that will 
contain the DOE SNF. 
The inventory estimates given in Source Term Estimates for DOE Spent Nuclear Fuels (DOE 
2003a) provide both a nominal and a bounding radionuclide inventory estimate for each type of 
DOE SNF. Even though the bounding radionuclide inventory estimates were provided for the 
purpose of assessing preclosure risk associated with handling a worst-case canister, the values 
are used to represent a bounding inventory for assessing postclosure risk as well (BSC 2003d, 
Section 6.6.2). The nominal and bounding inventories per waste package for the weighted 
average of all DOE SNF waste reported by Source Term Estimates for DOE Spent Nuclear Fuels 
(DOE 2003a) were analyzed. 
A range of total DOE SNF inventory (0.62 to 2 times the nominal inventory) was established on 
the following basis: 
• The nominal inventory includes extremely conservative assumptions applied to the small 
percentage of fuel (0.31%) for which little information is available. These assumptions 
result in 38% of the total inventory in 0.31% of the fuel, resulting in a potential 
overestimation of the total curie inventory by about 38% (DOE 2003a, p. 39). 
• The best estimate inventory has a ratio of 0.62 to the nominal inventory (BSC 2003d, 
Section 6.6.2). The best estimate inventory is lower than the nominal through removal 
of the extreme conservatism applied to the small percentage of fuel (0.31%) with little 
information and in effect replacing these assumptions with the approximation that the 
0.31% of the fuel has the same inventory as the average of the remaining fuel (DOE 
2003a). 
• The bounding radionuclide inventory is 1 to 2 times the nominal inventory per waste 
package (BSC 2003d, Section 6.6.2). 
Taking the total DOE SNF inventory, including uncertainty, the inventory was applied across the 
expected number of DOE SNF canisters. 
Because of uncertainty in the loading of fuel into the DOE SNF canisters, the DOE SNF canister 
count range is between 2,500 and 5,000, with a best estimate of 3,607 (DOE 2003a; Luptak 
2003). Applying a factor of 3,607/5,000 (0.72) to the best estimate inventory (0.62 times the 
July 2004 2-13 No. 7: In-Package Environment Revision 1 
nominal) and a factor of 3,607/2,500 (1.44) to the maximum DOE SNF inventory (2 times the 
nominal) and 3,607/5,000 to the reasonable canister inventories results in a per canister inventory 
range of 0.45 to 2.9 times the nominal, with a best estimate of 0.62 times the nominal. This 
range is shown in Table 2-4. 
238 
Like commercial SNF, the uncertainties of the DOE SNF radionuclide inventories are correlated, 
and an uncertainty multiplier is defined to capture the uncertainty for all radionuclides except 
U. The inventory of 238U has much less relative uncertainty than the other radionuclides 
because it is present in the initial fuel and generally changes little during reactor operation. Only 
three points are available to define a probability distribution for the DOE SNF multiplier, a 
minimum value, a most likely value, and a maximum value. In this situation, the choice of a 
triangular distribution is reasonable. Thus the DOE SNF multiplier is defined as a triangular 
distribution, with a minimum of 0.45, most likely value of 0.62, and a maximum of 2.9. It is to 
be applied to the nominal values for DOE SNF grams per waste package in Table 2-3 for all 
isotopes except 238U. 
2.5 SUMMARY AND CONCLUSIONS 
The radionuclide inventory component defines the four waste forms (commercial SNF; naval 
spent fuel (which is modeled as commercial SNF); DOE SNF; and HLW glass) and the two 
representative types of waste packages (commercial SNF waste packages and the codisposal 
waste packages containing both DOE SNF and HLW glass) planned for disposal in the 
repository. In addition, of the more than 100 radionuclides potentially present in the repository, 
28 radionuclides have been identified as being important to dose during the 10,000-year 
regulatory period. The initial inventory of these 28 radionuclides for the three waste forms 
modeled in TSPA-LA is given, as is the uncertainty associated with that inventory. The 
inventory of commercial SNF, which is the most abundant waste type to be disposed of in the 
2.4.3 High-Level Radioactive Waste Glass 
The uncertainty in the repository average inventories was investigated by comparing three 
radionuclide loadings and three canister-filling cases (BSC 2003d, Section 6.6.3). As was the 
case with commercial SNF and DOE SNF, the uncertainties in HLW radionuclide inventories are 
dependent on the same factors for HLW, in this case radionuclide loading per canister. 
Therefore, an uncertainty multiplier is again used to represent the uncertainty in the HLW 
inventory. With the current information, the nominal values reported in Table 2-4 are the most 
likely values, and thus the most likely value for the uncertainty multiplier is 1 (BSC 2003d, 
Section 6.6.3). The minimum number of waste packages is 0.70 times the central value, and the 
maximum number of canisters is 1.3 times the central value. Because of the uncertainty in 
possible new vitrified waste forms at the Savannah River Site with loading up to 50% higher, it 
is prudent to overestimate the upper limit. A maximum loading of 55% was chosen to provide 
margin, which corresponds to a ratio of 1.5 (BSC 2003d, Section 6.6.3). With the most likely 
maximum and minimum defined, a triangular distribution is chosen for the HLW uncertainty 
multiplier for use by the TSPA-LA model. This multiplier is to be applied to the nominal HLW 
inventories shown in Table 2-3 for all isotopes (including 238U). The uncertainty multiplier has a 
triangular distribution with a minimum of 0.70, most likely value of 1, and a maximum of 1.5 as 
shown in Table 2-4. 
2-14 July 2004 No. 7: In-Package Environment Revision 1 
repository, has the least uncertainty while the inventory of DOE SNF, which is the least 
abundant waste type to be disposed of in the repository, has the most uncertainty. Because the 
TSPA-LA model can simulate the degradation and failure of individual waste packages, the 
radionuclide inventories are provided on a grams-per-package basis. 
July 2004 2-15 No. 7: In-Package Environment INTENTIONALLY LEFT BLANK 
2-16 No. 7: In-Package Environment 
Revision 1 
July 2004 Revision 1 
3.1 ASSUMPTIONS 
As discussed in Section 2, two general types of waste packages are considered as representative 
of all waste packages: commercial SNF and codisposal waste packages containing both DOE 
SNF and HLW glass. For purposes of modeling in-package chemistry, the latter is represented 
by a waste package containing two N Reactor SNF canisters and two HLW canisters (BSC 
2004b, Section 6.3.3). N Reactor SNF was selected to represent the DOE SNF in the codisposal 
waste package because examination of the dissolution rate literature for other DOE SNF waste 
forms (DOE 2002b, Section 6) indicates that the degradation rate of N Reactor fuel generally 
exceeds that of the other types of DOE SNF and because N Reactor SNF comprises 
approximately 85% of the total metric tons of heavy metal of DOE SNF (DOE 2002b, 
Appendix D). 
The in-package chemistry model simulates chemical interactions of water with the waste 
package materials and waste forms for both types of waste packages under different physical, 
hydrologic, and chemical conditions. The simulations are performed with the reaction-path code 
EQ3/6 (Wolery 1992; Wolery and Daveler 1992), with the assumption that the water will always 
maintain equilibrium with precipitated secondary mineral phases as waste package components 
slowly dissolve. The outer shell of the waste package (Alloy 22) is assumed to be inert because 
of its extremely slow corrosion rate (CRWMS M&O 2000c, p. 109, Section 6.9.1). The 
calculation also assumes that oxygen is in equilibrium with the ambient atmosphere outside of 
the waste package. Therefore, the fugacity of O2 is set to the atmospheric value, 0.2 atm (BSC 
2004b, Section 6.3.1). Localized reducing conditions may occur as corrosion products limit the 
access of oxygen to metal surfaces. However, lower redox conditions would favor substantially 
lower radionuclide solubilities. 
Two different water ingress models are considered (BSC 2004b, Section 1): the no-drip model, 
which represents scenarios in which advection through the waste package does not occur, and the 
seepage-dripping model, which represents scenarios in which water can flow through the waste 
3. IN-PACKAGE CHEMISTRY 
Two sets of calculations were performed to estimate the chemistry of fluids reacting with SNF: 
one set represents conditions expected under the nominal and seismic scenario classes; the other 
set represents conditions expected under the igneous intrusion scenario. The bulk of this section 
describes the in-package chemistry model for the nominal and seismic scenario classes. 
Section 3.6 presents the in-package chemistry model for the igneous intrusion scenario. 
The in-package chemistry model simulates the chemistry of water as it reacts with waste package 
components and waste forms inside failed waste packages. The primary input parameters for the 
in-package chemistry model include the chemistry of seepage waters, the compositions of waste 
package materials and waste forms, and the corrosion rates of waste package components. The 
outputs of the model (which include pH, ionic strength, fluoride concentration, and total 
carbonate) are used, either directly or indirectly, as inputs to the models that evaluate dissolved 
concentrations of radionuclides, commercial SNF matrix degradation, HLW glass degradation, 
and colloid stability in the TSPA-LA calculations. The material presented in this section is 
primarily drawn from In-Package Chemistry Abstraction (BSC 2004b). 
3-1 July 2004 No. 7: In-Package Environment Revision 1 
package. In the no-drip model, water enters a failed waste package by vapor diffusion and 
subsequently reacts with the package materials and waste forms. The solutes exit the package by 
diffusion only. The evolution and the uncertainty range of in-package chemistry were simulated 
using individual waste package components or the combinations of these components as 
reactants, based on the consideration that water mixing may not fully occur in the waste package 
due to lack of advective flow. In the seepage-dripping model, seepage water drips into a waste 
package and continuously exits the package by advection after reacting with the package 
contents. In this case, the in-package chemistry is simulated by assuming that solutions reacting 
with various waste package components will be well mixed during the percolation inside a waste 
package. 
One of the model boundary conditions is that a continuous water film forms over the waste 
package components. The thickness of the film is fixed to be 1 mm for commercial SNF and 
2 mm for codisposal waste packages in the no-drip model, and 2.5 mm for commercial SNF and 
3.5 mm for codisposal waste packages in the seepage-dripping model. The basis for these 
thicknesses is that they represent the minimum values required by the EQ3/6 code to complete 
the entire suite of simulations. They are also reasonable separation half-distances for the 
situation being modeled; with degradation and collapse of basket materials, the interior of the 
waste form will consist of loosely compacted fuel and basket elements separated by millimeterto 
centimeter-size apertures. 
In the current model, no water evaporation is allowed to occur inside the waste package (BSC 
2004b, Section 6.3.2). This assumption may lead to a lower ionic strength of in-package water. 
As discussed in Section 7, a high ionic strength decreases colloid stability; thus, a lower ionic 
strength could allow more radionuclides to escape from the waste package in colloidal form 
(BSC 2003h, Figures 4 and 7). 
3.2 RELEVANT PROCESSES 
3.2.1 Water Fluxes 
In the no-drip model, it is assumed that water will form a continuous water film on waste 
package components. However, in reality, this may not occur because water condensation can 
only be possible when the waste package temperature is equal to or below the ambient 
temperature. Given the fact that the interior of a waste package is the source of heat and the 
waste package temperature after closure is maintained above 40°C during the 10,000-year 
regulatory period (BSC 2004e, Figure 6.5-3), the formation of a water film over waste package 
materials seems unlikely. As a simplifying assumption, this thermal effect is not considered in 
the in-package chemistry model. The diffusive flux of water vapor is calculated to be 0.8 L/yr 
(43.88 mol/yr) per waste package for isothermal conditions (BSC 2004f, Section 6.3.1). The 
no-drip model is simulated in EQ3/6 as a titration (Wolery and Daveler 1992). That is, water 
and the reactants are added to a reaction vessel at their respective kinetic rates or diffusion flux 
for water; products form and may redissolve, but no water exits the system (Wolery and Daveler 
1992, p. 42, Figure 3). 
In the seepage-dripping model, seepage drips onto the upper surface of the waste package and 
penetrates the waste package through an opening. The solution then flows through the waste 
July 2004 3-2 No. 7: In-Package Environment Revision 1 
package, forming a film on the interior waste package components, reacting, mixing, and 
transporting the dissolved material out of the waste package. There is no accumulation of water 
in the waste package. The water flux through the waste package is varied from 0.15 to 15.0 L/yr, 
representing low to moderate seepage rates (BSC 2003j, Figure 6.8-3). Use of a lower water flux 
increases the contact time of water with waste package components and thus allows the chemical 
reactions to progress further. For sensitivity analyses, a few simulations were performed with 
flux rates as high as 1,000 L/yr. In the seepage-dripping model, the solid-centered flow-through 
option in EQ6 is used. This option simulates a single-cell batch reactor. Seepage water �gflows�h 
into the cell at a specified rate while reactants are added to the cell at rates dictated by the 
degradation rates of their source materials. Water exits the cell at the same rate as it enters. 
3.2.2 Kinetic Degradation of Package Components 
The rate at which waste package components degrade, together with the rate of water influx, 
determine the evolution of in-package water chemistry. The degradation rate of commercial 
SNF is formulated as a function of pH and dissolved carbonate concentration, while the rate of 
HLW dissolution is described as a function of pH only, based on transition state theory and 
experimental measurements (see Section 5). For the corrosion of the codisposed N Reactor SNF 
and the metal alloys in the waste package, constant reaction rates are used. The reaction rates 
also depend on the surface areas of waste package components exposed, which have been 
estimated based on the configurations and dimensions of waste packages (BSC 2004b, Sections 
6.5.2 to 6.5.4). The surface areas are assumed to remain constant during degradation. This 
assumption is reasonable because the surface area would decrease as the quantity of a remaining 
reactant decreases while, on the other hand, it could also increase due to grain disaggregation. 
For the no-drip model, the calculations have been performed for the following percentages of 
surface area exposure: 10% for commercial SNF, 100% for HLW glass, and 50% for N Reactor 
SNF. For the seepage-dripping model, the following percentages are considered: 1%, 10%, and 
100% for commercial SNF, and 100% for both HLW and N Reactor SNF. These values were 
chosen only to parameterize the impact of coverage on effluent chemistry. The time-dependent 
percentage of commercial SNF surface area exposed as cladding splits (discussed in Section 4) 
and the surface areas of commercial SNF and high-level radioactive waste used in the waste 
form degradation models (discussed in Section 5) are not input to the in-package chemistry 
model when the model is implemented in TSPA-LA. 
(Eq. 3-1) 
(Eq. 3-2) 
3.2.3 Potential Generation of Alkaline or Acid Waters 
The waste package components could have a significant impact on the predicted in-package pH. 
The dissolution of high-level radioactive waste glass, which contains high concentrations of 
sodium and potassium, could generate an alkaline condition by releasing alkalis (Na+, K+, or 
Ca2+) into solution: 
Glass.Na+ + H+ �¨ Na+ + Glass.H+ 
In contrast, the Carbon Steel Type A516, which contains elemental sulfur, is a strong acid 
generator: 
Metal.S + H2O + 3/2O2 �¨ SO4
2. + 2H+ + Metal+ 
3-3 July 2004 No. 7: In-Package Environment Revision 1 
Similarly, the Stainless Steel Type 304L, also containing elemental sulfur, could also potentially 
generate an acid solution. However, its effect is expected to be much smaller than Carbon Steel 
Type A516, because of its very slow corrosion rate. The pH resulting from these two reactions 
will be altered by other chemical reactions such as surface complexation on corrosion products, 
leading to a near neutral pH range. 
3.2.4 Surface Complexation 
Steel corrosion, which is rapid relative to fuel oxidation, will result in a large accumulation of 
ferric (hydr)oxide corrosion products inside the waste package. It is estimated that 10 to 
40 moles of ferric (hydr)oxides (ferrihydrite, goethite, or hematite) are potentially able to form 
per liter of void space due to the degradation of low carbon and stainless steel (BSC 2004b, 
Section 6.3.3). Roughly half appears in the first 100 years after package failure due to the 
degradation of Carbon Steel Type A516. From 0.2% to 90% of precipitated iron(III) is able to 
react with solutions. This is equivalent to 0.02 to 40 moles of pH-buffering sites available per 
liter of water. Such a large buffering capacity will be a dominant factor controlling in-package 
chemistry. Given the fact that Carbon Steel Type A516 contains only trace amounts of sulfur 
(approximately 0.035 wt %) (Table 3-1), the proton release from reaction (Equation 3-2) will be 
overwhelmed by surface complexation on the corrosion products. Since the isoelectric point of 
ferric (hydr)oxides is approximately pH 7 to 8 (Davis and Kent 1990, Table 4, p. 210), the 
surfaces of these solids will acquire protons at pH less than 7 and lose them to solution at pH 
greater than 8, in effect limiting the pH range to near neutral. 
July 2004 3-4 No. 7: In-Package Environment Sulfur 
Silicon 
Nickel 
Cobalt 
Iron 
Boron 
Zinc 
Element 
Carbon 
Manganese 
Phosphorus 
Chromium 
Molybdenum 
Nitrogen 
Copper 
Magnesium 
Titanium 
Aluminum 
Total 
Source: BSC 2004b, Table 8. 
100.00 
NOTE: a A nickel alloy interspersed with gadolinium (UNS N06464) is recommended to replace Neutronit as the 
absorber material. The effects of this change on in-package chemistry should be negligible because 
deleting the Neutronit does not significantly change the quantity of available stainless steel; the low 
corrosion rates of the nickel alloy relative to stainless steel will result in slightly higher in-package pH 
(i.e., less aggressive), and gadolinium will form phosphates that are fairly insoluble (Loros and Williams 
2004). 
100.00 
3.2.5 Precipitation and Dissolution of Uranium Minerals 
The uranium mineral, schoepite, together with iron corrosion products, dominates the 
composition of the alteration products from 500 years and beyond (BSC 2004b, Section 6.8.2). 
Whether schoepite dissolution produces or consumes hydrogen ions depends on the pH of the 
solution: 
(Eq. 3-3) 
(Eq. 3-4) 
UO3¡¤2H2O + 2H+ ¡ê UO2
+2 + 3H2O for pH ¡Â 5.5 
These two reactions constitute a negative feedback that resists pH excursions to high or low pH, 
and ultimately constrains pH values near the schoepite solubility minimum pH 6.5 to 7, 
No. 7: In-Package Environment 
Table 3-1. Composition of Steel and Aluminum Alloys 
Carbon 
Steel Type 
A516 
(wt %) 
0.28 
1.045 
0.035 
0.035 
0.29 
. 
. 
. 
. 
. 
98.3 
. 
. 
. 
. 
. 
. 
Neutronita 
(wt %) 
0.04 
. 
. 
. 
. 
18.5 
13 
0.2 
2.2 
. 
64.82 
1.245 
. 
. 
. 
. 
. 
100.00 
UO3¡¤2H2O + 3HCO3 
. ¡ê UO2(CO3)3 
.4 + H+ + 3H2O for pH > 7-7.5 
SS31600 
(wt %) 
0.02 
2.00 
0.045 
0.03 
0.75 
17.00 
12.00 
. 
2.50 
0.08 
65.58 
. 
. 
. 
. 
. 
. 
Al-6061 
(wt %) 
. 
0.15 
. 
. 
0.60 
0.195 
. 
. 
. 
. 
0.7 
. 
0.25 
0.275 
1.0 
0.15 
96.68 
100.00 100.00 
3-5 
Revision 1 
Stainless 
Steel Type 
304L 
(wt %) 
0.03 
2.00 
0.05 
0.03 
0.75 
19.00 
10.00 
. 
. 
0.10 
68.05 
. 
. 
. 
. 
. 
. 
Al-1100 
(wt %) 
. 
. 
. 
. 
0.45 
. 
. 
. 
. 
. 
0.50 
. 
. 
0.05 
. 
. 
99.00 
100.00 
July 2004 Revision 1 
depending on the ambient carbon dioxide fugacity. The carbon dioxide fugacity determines the 
total concentration of aqueous carbonate species (see Section 3.5.3). 
3.2.6 Effect of Radiolysis 
When radiation passes through a material, some energy is deposited in the medium and chemical 
reactions can occur from the local deposition of energy (radiolysis). When gamma and fast 
neutron energy pass through moist air, nitric acid is produced. Hydrogen peroxide is generated 
from the radiolysis of water. Thus, radiolysis could potentially influence the results of the 
in-package chemistry model. However, EQ3/6 simulations show that the amount of acid 
produced from radiolysis has an insignificant effect on in-package chemistry (BSC 2004b, 
Attachment III). 
3.2.7 Effect of Temperature 
Most EQ3/6 simulations were run at either 25°C or 50°C. However, a few simulations were run 
at higher temperatures to investigate the effect of temperature on pH and ionic strength (BSC 
2004b, Section 6.7.5). The results show that as temperature increases, pH tends to decrease and 
ionic strength remains largely unaffected. The reason for the decrease in pH is related to the 
increased dissociation of the water molecule as temperature rises. The effect of temperature was 
included in the pH abstraction used in TSPA-LA, as described in Section 3.5.1. 
3.3 SOURCE OF DATA 
The primary input parameters for the in-package chemistry model include the chemistry of 
seepage waters, the compositions of waste package materials and waste forms, and the corrosion 
rates of waste package components (BSC 2004b, Section 4). For reference, some of these 
parameters are listed in Tables 3-1 through 3-4. Table 3-4 gives experimental corrosion rates of 
the metal alloys in the waste package. The corrosion rate data in Table 3-4 were collected under 
a variety of conditions and provide the basis for the metal alloy corrosion rates used in the inpackage 
chemistry abstraction and shown in the last column of Table 3-4. 
July 2004 3-6 No. 7: In-Package Environment Table 3-2. Input Water Composition for Model Calculation 
Units 
mg/L 
mg/L 
mg/L 
mg/L 
mg/L 
mg/L 
mg/L 
mg/L 
mg/L 
mg/L 
mg/L 
pH 
Source: BSC 2004b, Table 2. 
Parameter 
Ca 
Mg 
Na 
K 
Si 
SiO2 
NO3 
HCO3 
Cl 
F 
SO4 
pH 
NOTE: 
c 
aSample: ECRB-SYS-CS1000/7.3-7.7/UC; DTN: GS020408312272.003. 
bSample: ECRB-SYS-CS2000/16.3-16.5/UC; DTN: GS020408312272.003. 
DTN: MO0006J13WTRCM.000. 
d Harrar et al. 1990, pp. 4–9. 
Calcium Pore Water 
ECRB-SYS-CS1000/ 
7.3-7.7/UCa 
94 
18.1 
39 
7.6 
N/A 
42 
2.6 
397 
21 
3.4 
36 
7.6 
No. 7: In-Package Environment 
Sodium Pore Water 
ECRB-SYS-CS2000/ 
16.3-16.5/UCb 
81 
3.3 
120 
6.1 
N/A 
42 
0.41 
362 
24 
6 
31 
7.4 
3-7 
Revision 1 
Well J-13c 
13.0 
2.01 
45.8 
5.04 
28.5 
N/A 
8.78 
Calculated from 
charge balance 
7.14 
2.18 
18.4 
7d 
July 2004 Table 3-3. Chemical Composition of Commercial Spent Nuclear Fuel, N Reactor Fuel, and High-Level 
Radioactive Waste Glass 
Element 
Uranium 
Neptunium 
Plutonium 
Zirconium 
Molybdenum 
Technetium 
Ruthenium 
Cesium 
Barium 
Gadolinium 
Oxygen 
Aluminum 
Sulfur 
Calcium 
Phosphorus 
Silicon 
Boron 
Fluorine 
Iron 
Potassium 
Magnesium 
Sodium 
Source: BSC 2004b, Tables 10 and 11. 
No. 7: In-Package Environment 
Commercial SNF 
(mol/100 g) 
0.3617 
0.0009 
0.0027 
0.0005 
0.0009 
0.0008 
0.0020 
0.0013 
0.0010 
0.0035 
0.7385 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
N Reactor Fuel 
(mol/100 g) 
0.42 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
N/A 
3-8 
Revision 1 
High-Level 
Radioactive Waste 
Glass 
(mol/100 g) 
0.00782 
0 
0 
0 
0 
0 
0 
0 
0.00108 
0 
2.70 
0.0863 
0.00401 
0.0162 
0.000489 
0.776 
0.291 
0.00166 
0.172 
0.0751 
0.0333 
0.577 
July 2004 Metal 
Carbon Steel 
Type A516 
(times less than 
0.53 years) 
Carbon Steel 
Type A516 
(times greater 
than 1.0 years) 
Neutronit 
Aluminum Alloy 
Stainless Steel 
Type 316 
Stainless Steel 
Type 304L 
Source: BSC 2004b, Table 9; BSC 2004g, Table 7-4. 
Min. 
78.71 
58.08 
50.25 
7.39 
65.77 
29.53 
6.77 
3.69 
0.001 
0.025 
1.810 
0.40 
0.12 
0.037 
0.001 
1.588 
0.660 
NOTE: ó = standard deviation; SDW = simulated dilute water; SCW = simulated concentrated water. See note in 
Table 3-1. 
Rate Used in the 
In-Package 
Chemistry Model 
(µm/yr) 
72 
0.1 
3.0 
0.1 
0.1 
3.3.1 Initial Water Compositions 
For the seepage-dripping model, three water compositions have been used for the initial 
conditions (BSC 2004b, Table 2), as shown in Table 3-2: (1) a calcium-dominated pore water, 
(2) a sodium-dominated pore water, and (3) a J-13 well water. The calcium- and 
sodium-dominated pore water compositions were used because they were obtained from core 
samples near the repository. The decision to use these water compositions was based on several 
lines of reasoning. The J-13 well water composition was used for comparison purposes (i.e., to 
maintain continuity between the current work and past in-package chemistry analyses). The 
calcium and sodium pore water compositions, corresponding to pore waters “W5” and “W4,” 
respectively in Drift-Scale Coupled Processes (DST and THC Seepage) Models (BSC 2004e), 
are used because they were obtained from core samples proximal to the repository. These two 
pore waters represent possible in situ water compositions in the unsaturated repository horizon. 
Pore water compositions can also be perturbed thermally during the postclosure time period. 
However, a sensitivity analysis has shown that this perturbation is negligible (BSC 2004b, 
No. 7: In-Package Environment 
Table 3-4. Metal Alloy Corrosion Rates 
0.0014 14.787 0.7362 
Median Mean 
130.70 101.95 102.71 
81.14 77.05 
68.08 62.77 
12.78 12.42 
77.43 74.56 
51.80 48.70 
10.61 10.83 
6.84 6.75 
0.004 0.003 
0.206 0.203 
11.060 10.190 7.38 
12.95 9.50 
9.69 4.76 
0.0083 0.0136 0.003 
0.248 0.229 
1.939 
0.214 0.1285 
11.441 11.134 5.08 
5.816 2.03 
Freshwater (29.5°C) 0.0007 0.0475 
Max. 
130.02 
104.20 
22.06 
106.93 
88.68 
14.36 
9.35 
0.011 
0.330 
29.220 
36.93 
110.91 
0.51 
1.570 
39.147 
15.900 
Condition 
SDW (60°C) 
SDW (90°C) 
SCW (60°C) 
SCW (90°C) 
SDW (60°C) 
SDW (90°C) 
SCW (90°C) 
SCW (90°C) 
Freshwater (29.5°C) 
Freshwater 
(50°C to 100°C) 
Saltwater (26.7°C) 
Freshwater 
Saltwater 
Freshwater 
(50°C to 100°C) 
Saltwater (26.7°C) 
Freshwater 
(25°C to 100°C) 
Saltwater (26.7°C) 
Saltwater (90°C) 
Corrosion rate (µm/yr) 
3-9 
Revision 1 
ó 
12.37 
15.13 
14.17 
3.77 
8.83 
12.99 
2.02 
1.25 
0.004 
0.088 
10.84 
15.34 
0.146 
3.346 
0.298 
5.953 
July 2004 Revision 1 
Figure 14). For the no-drip model, the initial water is assumed to be pure water in equilibrium 
with the ambient fugacities of CO2 and O2 (BSC 2004b, Section 6.3). 
3.3.2 Compositions of Waste Package Components 
The chemical composition of each component is listed in Table 3-3. Table 3-3 summarizes the 
composition of commercial SNF, N Reactor fuel, and HLW glass. As discussed in Section 2, the 
commercial SNF is an oxide fuel while the N Reactor SNF is a uranium metal fuel. 
3.3.3 Corrosion Rates 
Metal corrosion rates are used to estimate the mean lifetime of the waste form internals and 
allow prediction of the time needed for structural collapse to occur. Metal corrosion proceeds 
initially by the direct oxidation of reduced constituents (iron, aluminum, chromium, etc.) to form 
a coating of metal hydroxides that can slow the further access of water and oxygen to the 
dissolving metal surface. Consequently, absolute rates of corrosion depend on both the oxidation 
step and the stability and chemical reactivity of the corrosion product layer. Solution chemistries 
that solubilize the latter, for example, can accelerate corrosion. Corrosion rates are important to 
the in-package chemistry model because they allow some predictions to be made of the rate of 
metal hydroxide accumulation, which is important for pH control, radionuclide sorption, and 
potential colloid formation. The corrosion rates of metal alloys are given in Table 3-4. These 
rates will encompass various aqueous parameters such as temperature (up to 100°C), water type 
(i.e., fresh versus saline), and pH. The corrosion rates of the different waste forms can be found 
in Section 5 of this document. Note that the corrosion rate of Carbon Steel Type A516 is about 
3 orders of magnitude higher than that of Stainless Steel Type 304L. Therefore, even though 
both materials contain a similar sulfur content, their impacts on in-package pH will be quite 
different, as discussed in Section 3.2.3. 
3.4 EVOLUTION OF IN-PACKAGE CHEMISTRY 
The temporal evolution of in-package chemistry is calculated by assigning kinetic rates to waste 
package component degradation. Note that the starting time for the in-package chemistry model 
refers to the time at which waste packages are breached and sufficient water becomes available 
for chemical reactions. As discussed in Section 3.2, the in-package chemistry is controlled not 
only by waste and waste package corrosion and the associated mineral precipitation or 
dissolution, but also by surface complexation on corrosion products. The EQ3/6 code does not 
model interfacial chemical processes such as surface complexation. Thus, in-package conditions 
have been modeled in two steps. First, EQ3/6 calculations were made simulating the evolution 
of these conditions without considering the effects of surface complexation reactions for 
20,000 years following the start of reactions in the package (which is significantly beyond the 
10,000-year regulatory period). Second, the effects of surface complexation reactions on the 
in-package pH are modeled. The surface sites on which these reactions take place will be 
available for at least several hundred years, as long as a significant amount of steel degradation is 
occurring. During this period, surface complexation rather than the reactions modeled by EQ3/6 
will likely control the in-package pH (BSC 2004b, Section 6.3.3). 
3-10 July 2004 No. 7: In-Package Environment Revision 1 
The pHs predicted inside the waste form tend to be near neutral indicating that the pH-buffering 
ability of corrosion product surface complexation and schoepite dissolution will limit pH 
excursions. The isoelectric point of ferric corrosion products and the solubility minimum of 
schoepite are both near neutral. Equilibration of acidic or alkaline fluids with ferric corrosion 
products and schoepite consequently limits pHs to between approximately 4.5 and 8. Increasing 
carbon dioxide levels causes a slight shift of pHs to lower values while favoring the formation of 
carbonate-containing surface complexes at the corrosion product-solution interface and dissolved 
uranyl-carbonate species. The pH abstractions capture the net effect of these processes (BSC 
2004b, Section 6.8.2). 
3.4.1.1 No-Drip Model 
Figure 3-1 displays the EQ3/6 simulation results for individual commercial SNF components at 
25°C (BSC 2004b, Figure 1). The figure provides information on how each waste package 
component contributes to the in-package pH and shows the upper and lower pH limits for a 
commercial SNF package without considering the effects of surface complexation reactions. 
Note that the horizontal axis is in terms of moles of material dissolved independent of time. The 
low pH resulting from Carbon Steel Type A516 corrosion is due to the combination of oxidation 
of elemental sulfur, the large quantity of Carbon Steel Type A516, and its high corrosion rate 
relative to the other waste package components. Both the Al-6061 and the commercial SNF 
have neutral to slightly basic pH profiles, while the S31600 and Neutronit (borated Carbon Steel 
Type 316) have slightly acidic pH profiles. 
The simulations have also been performed for multicomponent ensembles and resulted in a 
similar pH range (BSC 2004b, Figure 2). The simulations show that if it were not for the surface 
complexation reactions, the early-time pH (less than 100 years) will be controlled by Carbon 
Steel Type A516 dissolution, and for 100 to 500 or 1,000 years the pH would be controlled by 
Al-6061 dissolution. At times greater than 500 to 1,000 years, when surface complexation 
reactions may no longer be effective, the pH will be controlled by equilibrium with the corrosion 
products. 
3.4.1 Commercial Spent Nuclear Fuel Waste Packages 
3-11 July 2004 No. 7: In-Package Environment Revision 1 
Source: DTN: MO0403SPAIPCHM.004, CSNF_singlereact_NDM.xls. 
3.4.1.2 Seepage-Dripping Model 
The evolution of in-package chemistry for the median water flux and 10% fuel exposure at 25°C 
predicted by EQ3/6 is shown in Figure 3-2. Although nominal temperatures in the repository 
will exceed 40°C for the first 10,000 years, 25°C runs were used initially to identify the relevant 
chemical processes and trends as thermodynamic and kinetic data are better known at the slightly 
lower temperature. It can be seen that the compositions of seepage water have virtually no effect 
on the pH evolution. The solution pH will be quickly overwhelmed by waste package corrosion 
over the first 10 years. Low pH values modeled without considering the effects of surface 
complexation reactions result from the relatively fast corrosion of a large quantity of Carbon 
Steel Type A516 corrosion (Figure 3-2). As shown in the figure, however, this pH will be 
buffered by the corrosion of other waste package components after a few hundred years. The 
commercial SNF is predicted to completely degrade over approximately 5,000 years. 
Figure 3-1. pH Calculated as a Function of Moles of Reactant Dissolved in Kilograms of Water for 
Commercial Spent Nuclear Fuel Single Component Degradation under No-Drip Conditions 
July 2004 3-12 No. 7: In-Package Environment Revision 1 
NOTE: C22C25B = Ca-pore water, C22J25B = J-13 pore water, and C22N25B = Na-pore water. 
Source: DTN: MO0403SPAIPCHM.004, CSNF_SDM_25_rev03.xls. 
3.4.2 Codisposal Waste Packages 
3.4.2.1 No-Drip Model 
Figure 3-3 displays the 25°C pH profiles for individual codisposal waste package components as 
calculated by EQ3/6. Again, note that the horizontal axis is in terms of moles of material 
dissolved independent of time. The codisposal waste package displays a much wider variation in 
pH compared to the commercial SNF package (Figure 3-1). This is because the HLW generates 
high pH conditions due to its high concentrations of sodium and potassium. The Al-1100, 
S31600, and N Reactor fall in the middle of the pH range at approximately 5.7 to 6.2. The 
corrosion of Carbon Steel Type A516 and Stainless Steel Type 304L corrosion leads to pH 
values of 1.5 to 2, as shown in Figure 3-3. However, over the first few hundred years this low 
pH will be buffered by surface complexation reactions of the corrosion products (BSC 2004b, 
Section 6.8.2). Furthermore, over longer time periods, results of the codisposal waste package 
multicomponent ensemble runs show that this low pH will be buffered by either HLW glass or 
N Reactor fuel, since the TSPA-LA model does not take cladding credit for N Reactor fuel or the 
glass pour canisters (BSC 2004b, Section 6.6.2.1, Figure 9). 
Figure 3-2. Evolution of pH and Reactants for Commercial Spent Nuclear Fuel Waste Packages under 
1.5-L/yr Seepage-Dripping Conditions and 10% Fuel Exposure at 25°C 
July 2004 3-13 No. 7: In-Package Environment Revision 1 
Source: DTN: MO0403SPAIPCHM.004, CD_singlereact_NDM.xls. 
3.4.2.2 Seepage-Dripping Model 
Figure 3-4 displays the 25°C evolution of pH and reactants for codisposal waste packages under 
seepage conditions. The pH profiles converge early in the simulations indicating that the model 
pH response is insensitive to starting water composition, which is similar to the pH profile 
behavior of the commercial SNF seepage-dripping model (Figure 3-2). The figure shows that 
the N Reactor fuel dissolves in just a few years. The rapid dissolution of the N Reactor fuel 
diminishes its contribution to the overall in-package chemistry. However, the resulting large 
quantity of schoepite may have the capacity to influence the in-package chemistry for an 
extended duration. The minimum pH for codisposal waste packages is approximately one unit 
higher that for commercial SNF under the same chemical and physical conditions (Figure 3-2). 
Another important feature of the pH profiles in Figure 3-4 is the absence of a period of sustained 
high pH (greater than 9) that might be expected from the dissolution of the HLW glass. The 
codisposed no-drip models, single reactant and ensemble, both predicted that high pH conditions 
are possible in the absence of an acid-producing reactant (i.e., steel alloy) (Figure 3-3). 
However, in the seepage-dripping model, seepage is allowed to “stream” through the waste 
package reacting with waste package components, and it is unlikely that seepage, which has 
reacted with HLW glass could exit a waste package without contacting a steel component along 
its flow path. 
Figure 3-3. pH Calculated as a Function of Moles of Reactant Dissolved in Kilograms of Water for 
Codisposed Single Component Degradation under No-Drip Conditions 
July 2004 3-14 No. 7: In-Package Environment Revision 1 
Source: DTN: MO0403SPAIPCHM.004, CDNR_SDM_25_rev03.xls. 
Seepage-Dripping Conditions and 10% Fuel Exposure at 25�‹C 
Figure 3-4. Evolution of pH and Reactants for Codisposal Waste Packages under 1.5-L/yr 
(Eq. 3-6) 
(Eq. 3-7) 
(Eq. 3-8) 
July 2004 
3.4.3 pH Regulated by Surface Complexation 
As discussed in Section 3.2.4 (also see BSC 2004b, Section 6.3), a large quantity of Fe 
oxyhydroxides formed from steel corrosion will be available for surface complexation reactions. 
The surface protonation and deprotonation reactions as described by (Dzombak and Morel 1990; 
Appelo et al. 2002) are: 
+ .>Fe-O-H + H+ K1 = [>Fe-O-H][H+]/[>Fe-O-H2
+] = 10.7.29 (Eq. 3-5) >Fe-OH2 
>Fe-OH .>Fe-O. + H+ K2 = [>Fe-O.][H+]/[>Fe-O-H] = 10.8.93 
where >Fe denotes sites exposed at the FeOOH-solution interface. Sorption of bicarbonate ions 
may also occur: 
>Fe-O-H + H+ + CO3 
2. .>Fe-HCO3 
. + H2O 
KCO3 = [>Fe-HCO3 
.]aH2O/[>Fe-OH][H+][ CO3
2.] = 1012.78 
3-15 No. 7: In-Package Environment Revision 1 
By accounting for the surface complexation effect, the Carbon Steel Type A516 dissolution 
reaction can be written as: 
Fe1.76Mn0.019C0.0233Si0.0103P0.00113S0.00109 + mH2O + fO2 ¡ú 
0.00109SO4
2. + 0.00113(H3PO4 + H2PO4 
. + HPO4
2.) + 0.0103H4SiO4 + 
2CO3 + HCO3 
.) + 0.019MnO2 + (1.76 . X) FeOOH + 0.0233(H 
X(>Fe-O-H + >Fe-O-H2
+ + >Fe-O. + >Fe-O-CO3 
.) + nH+. (Eq. 3-9) 
where X is the amount of Fe in FeOOH (moles) that is exposed to solution, mH2O is moles of 
water, and fO2 is fugacity of oxygen. 
Given a value of X = 0.176 (10% of Fe sites, a reasonable value for goethite) for dissolution of 
1 mol of Carbon Steel Type A516 into a near neutral solution: 
4 
(Eq. 3-10) 
n = 2[SO 2.] + [H2PO4 
.] + 2[HPO4
2.] + [HCO3 
.] + 
[>Fe-O.] + [>Fe-O-HCO3 
.] ¨C [>Fe-O-H2
+]. 
Similarly, charge balance for a solution in contact with carbon dioxide gas at 0.001 atm and into 
which 1 mol of Carbon Steel Type A516 were dissolved would be: 
(Eq. 3-11) 
(H+) + (>Fe-O-H2
+) = 2[SO4
2.] + [H2PO4 
.] + 2[HPO4
2.] + 
(OH.) + (HCO3 
.) + (>Fe-O.) +(>Fe-O-HCO3 
.). 
The two phosphate terms can be manipulated with the equilibrium constant for their conversion 
to give: 
[H2PO4 
.] = 0.00113/(1 + 10.7.2/[H+]) and [HPO4
2.] = 
0.00113(1 ¨C 1/(1 + 10.7.2/[H+])). (Eq. 3-12) 
Substituting these expressions, carbonate equilibrium constants, the surface equilibria above, and 
rearranging to solve for pH: 
(Eq. 3-13) 
(H+) = 0.00218 + 0.00113/(1 + 10.7.2/[H+]) + 
0.00226(1 ¨C 1/(1 + 10.7.2/[H+])) + 0.176[10.8.93/(H+) + 
pCO210.5.28/(H+) ¨C 107.29(H+)] + 10.14/(H+) + 10.7.82 fCO2/(H+) 
The equation above can be solved for pH as a function of fixed fugacity of CO2. For the relevant 
fugacity of CO2, the calculation shows that the surface complexation on the corrosion products 
of Carbon Steel Type A516 can sufficiently neutralize the acidic solution generated by sulfur 
oxidation and that the solution¡¯s pH will be buffered at 7.3 to 8.1. This will occur immediately 
after onset of A516 corrosion (that is within the first several decades after waste package 
breach). The effect of increasing carbon dioxide levels in the waste form is to lower the ambient 
pH. 
3.4.4 Upper pH Limit 
As discussed in Section 3.2.5, schoepite together with iron corrosion products dominates the 
composition of the alteration products from time periods of 500 years and beyond. Schoepite 
July 2004 3-16 No. 7: In-Package Environment dissolves at pH values below about 5.5 to consume protons and at pH values above about 7 to 
produce protons (Equations 3-1 to 3-2). These two reactions constitute a negative feedback that 
will constrain pH at 5.5 to 7.5. As Figure 3-3 shows, HLW glass reaction by itself could lead to 
pH values from 8 to 10. For such high pH values to develop, it would require an unlikely 
scenario in which water reacts with the glass without also reacting with schoepite inside the 
waste package. Thus, a “HLW glass only, no steel, no fuel” scenario that produces high pH 
values is highly unlikely to occur. The most reasonable high-end pH for waste packages 
containing HLW glass is therefore 6.5 to 7. 
3.5 IMPLEMENTATION FOR NOMINAL AND SEISMIC SCENARIOS 
3.5.1 Abstraction of pH 
3.5.1.1 Commercial Spent Nuclear Fuel Waste Package pH Abstraction 
During the period when surface complexation is implemented (0 to 600 years) and based on the 
results of the surface complexation analyses, the pH for a commercial SNF waste package for 
both the nondripping and the dripping models is constrained by the ranges defined as a function 
of the fugacity of carbon dioxide gas, as shown in Table 3-5. 
The minimum pH value, pH 4.5, was set based on a simulation at 90°C, indicating that at times 
greater than 600 years the pH stabilizes around 4.5. This simulation is shown in Figure 3-5 and 
represents a minimum pH based on the EQ6 results that do not include surface complexation to 
account for the possibility of localized areas of low pH due to the potential of heterogeneous 
flow through the waste package internals. The maximum pH values for the 0-to-600-year period 
are calculated from Equation 3-13. 
Table 3-5. Commercial Spent Nuclear Fuel Lookup Table of pH as a Function of Carbon Dioxide 
Fugacity 
Period (years) 
0 to 600 
minimum pH 
4.5 
4.5 
4.5 
4.5 
4.5 
4.5 
4.5 
4.5 
4.5 
log(fCO2) 
-5.0 
-4.5 
-4.0 
-3.5 
-3.0 
-2.5 
-2.0 
-1.5 
N/A 600 to 20,000 
Source: BSC 2004b, Table 66; DTN: MO0404SPAIPCHM.005. 
3-17 No. 7: In-Package Environment 
Revision 1 
maximum pH 
8.1 
8.1 
8.0 
7.9 
7.7 
7.5 
7.3 
7.0 
7.0 
July 2004 Revision 1 
Source: BSC 2004b, Figure 24. 
NOTE: C22C25 = 1.5-L/yr seepage dripping conditions, 10% fuel exposure, and calcium pore water at 25°C 
Figure 3-5. Effect of Increasing Temperature on the pH Profile for Commercial Spent Nuclear Fuel Waste 
Package 
During the period from 600 to 20,000 years, the minimum pH is set by the high temperature 
results shown in Figure 3-5 and the maximum by the 50-L/yr results shown in Figure 3-6. These 
values provide the maximum plausible range of pH. 
For implementation in TSPA, the 0- to 600-year pH distributions are set using the minimum and 
maximum pH values (Table 3-5) for each log(fCO2). Values for the latter are taken from the 
EBS physical and chemical environment model (BSC 2004h). Each distribution is uniform, and 
pHs for intermediate values of log(fCO2) are linearly interpolated between nearest neighbor 
values. No uncertainty term is added to the sampled value. The abstracted pH values are 
independent of both time and temperature, and are not correlated to the pH in codisposal waste 
packages. 
For the period from 600 to 20,000 years the pH should be uniformly sampled between pH 4.5 
and pH 7 with no additional uncertainty term added. There is no correlation between this pH 
abstraction and that for codisposal waste packages. 
July 2004 3-18 No. 7: In-Package Environment Revision 1 
Source: BSC 2004c, Figure 20. 
NOTE: Cases run using calcium–pore water, 10% fuel exposure, at 25°C. 
2 
3.5.1.2 Codisposal Waste Package Abstractions 
The abstraction for pH inside a failed codisposal waste package, for both the dripping and 
nondripping models, is given in Table 3-6, and the basis for the minimum and maximum values 
are as follows. When the water flux is less than 150 L/yr per waste package, the maximum pH 
inside a codisposal waste package will not exceed 7 independent of temperature and CO 
fugacity due to a variety of buffering reactions, for example, schoepite dissolution (BSC 2004b, 
Section 6.8.2). For water flux values greater than 150 L/yr per waste package, the maximum pH 
is 8 (BSC 2004b, Figure 22). The lower pH bound, pH = 4.5, was set by the high temperature 
analysis for both flux conditions (BSC 2004b, Figure 25). 
For the purposes of implementation in TSPA, a uniform sampling of the ranges given in 
Table 3-6 with no additional uncertainty will capture the proper pH behavior for nondripping and 
seepage dripping scenarios in a codisposal waste package. 
Table 3-6. Codisposal pH Abstraction 
Period (years) 
0 to 20,000 
Maximum 
pH 
7.0 
Minimum 
pH 
4.5 
Flux Condition 
(L/yr per waste package) 
Q<150 
8.0 4.5 
Source: BSC 2004b, Table 67. 
Figure 3-6. Effect of Water Flux on pH for a Commercial Spent Nuclear Fuel Waste Package 
July 2004 
Q>150 
3-19 No. 7: In-Package Environment Revision 1 
3.5.2 Ionic Strength Abstraction 
The ionic strength results predicted using the EQ6 model were based on layer thickness values 
adjusted to avoid exceeding the ionic strength limit of the aqueous model. Thus, the ionic 
strength boundary condition was set by the model itself. Ionic strength data are used by the 
colloid submodel to determine if the colloid stability threshold, 0.05 mol/kg ionic strength, has 
been exceeded. This threshold value is much below the upper limits of the aqueous model (not 
less than 1 mol/kg) so that use of the higher EQ6 calculated ionic strengths is acceptable. 
The ionic strengths predicted by EQ6 may be high because the underlying calculations do not 
account for removal of dissolved salts from solution by corrosion product sorption. However, 
the inability to predict the identities and abundances (hence site densities) of the corrosion 
products over long periods of time prevents estimation of the magnitude of this effect. For 
example, complete conversion of corrosion products to low site density hematite over time 
would minimize any change in the ionic strengths predicted by EQ6. The possibility of ionic 
strength lowering by sorption reactions is dealt with by setting the uncertainty range to favor 
lower ionic strengths. 
(Eq. 3-14) 
3.5.2.1 
where 
Ionic Strength for Commercial Spent Nuclear Fuel Waste Packages 
The ionic strength distribution for commercial SNF no-drip conditions is constrained from the 
ensemble calculations (BSC 2004b, Figure 28). In the TSPA-LA model, it is sampled from 
0.085 to 0.75 mol/kg with a loguniform probability distribution function. A loguniform 
distribution is used to envelop calculated ionic strengths that themselves vary by orders of 
magnitude. The uncertainty in ionic strength described below should be added to the sampled 
value. 
For a seepage-dripping condition, the ionic strength is calculated as 
y = aeb log f 
y = ionic strength (mol/L) 
a = constant (unitless) 
b = constant (unitless) 
f = seepage flux (L/yr). 
The values of a and b used in Equation 3-14 were calculated from a curve-fitting procedure and 
are shown in Table 3-7, along with the appropriate temperature, flux limits, and regression 
coefficient for the curve fit. Values of ionic strength are either interpolated or linearly 
extrapolated from the 25°C and 50°C values to other temperatures from 25°C to 99°C (the limits 
of the in-package chemistry abstraction). As the seepage rate increases, the ionic strength inside 
the package will approach the values of initial waters. If the flux is less than the minimum, then 
the ionic strength should be evaluated using the lower flux limit. Likewise, if the flux is greater 
than the upper limit, then the ionic strength should be evaluated using the upper limit value. 
3-20 July 2004 No. 7: In-Package Environment Table 3-7. Commercial Spent Nuclear Fuel Seepage Dripping Model Ionic Strength Abstraction 
b (unitless) 
-2.1652 
a (unitless) 
0.4755 
Temperature (°C) 
25 
-2.2424 0.4472 50 
Flux Limits Per Waste Package 
(L/yr) 
Lower: 0.20 
Upper: 185 
Lower: 0.16 
Upper: 145 
Source: BSC 2004b, Tables 70 to 73. 
3.5.2.2 Ionic Strength for Codisposal Waste Packages 
The ionic strength distribution for codisposal waste packages under no-drip conditions is 
constrained from the ensemble calculations (BSC 2004b, Figure 32). In the TSPA-LA model, it 
is sampled from 0.0032 to 0.75 mol/kg with a cumulative probability distribution shown in 
Table 3-8. The distribution in Table 3-8 is derived from the distribution of ionic strengths 
calculated in the reaction path runs. 
Table 3-8. Codisposal Waste Package No-Drip Model Ionic Strength Cumulative Distribution 
Time Period after Waste Package Breach 
(years) 
0 to 20,000 
Ionic Strength 
(mol/kg) 
3.2 × 10-3 
0.05 
0.75 
Source: BSC 2004b, Table 74. DTN: MO0404SPAIPCHM.005. 
For dripping conditions, the ionic strength is a function of seepage rate as shown in 
Equation 3-14, as it is for commercial SNF waste packages. The values of a and b used in 
Equation 3-14 were calculated from a curve-fitting procedure and are shown in Table 3-9, along 
with the appropriate temperature, flux limits, and regression coefficient for the curve fit. Values 
of ionic strength are either interpolated or extrapolated to other temperatures. For example, 
while 90°C values are not shown, they can be readily extrapolated from the 25°C and 50°C 
values. As the seepage rate increases, the ionic strength inside the package will approach the 
values of initial waters. If the flux is less than the minimum, then the ionic strength should be 
evaluated using the lower flux limit. Likewise, if the flux is greater than the upper limit, then the 
ionic strength should be evaluated using the upper limit value. 
Table 3-9. Codisposal Waste Package Seepage Dripping Model Ionic Strength Abstraction 
b 
-2.1306 
a 
0.5201 
Temperature (°C) 
25 
0.4324 50 
Flux Limits Per Waste Package 
(L/yr) 
Lower: 0.25 
Upper: 180 
Lower: 0.18 
Upper: 180 
Source: BSC 2004b, Tables 75 through 78. 
-2.1369 
3-21 No. 7: In-Package Environment 
Revision 1 
R2 
0.9979 
0.9971 
Probability 
0.0 
0.13 
1.0 
R2 
0.9993 
0.9993 
July 2004 Revision 1 
3.5.2.3 
(Eq. 3-15) 
Uncertainty in Ionic Strength 
The variability of ionic strength is illustrated in Figure 3-7. For both the commercial SNF and 
codisposal waste package seepage-dripping models and no-dripping models at temperatures 
ranging from 25°C to 99°C (i.e., the temperature boundaries of the in-package chemistry 
abstraction) a uniformly sampled uncertainty value from the range of -0.8 to 0.4 multiplied by 
the sampled value is added to the sampled value. This range accounts for uncertainties in carbon 
dioxide levels as well as the possibility of lower ionic strengths as a result of surface 
complexation effects. Mathematically, the uncertainty for dripping and nondripping cases is 
implemented as follows: 
Ionic strength = ISsampled + (Uncertaintysampled × ISsampled) 
Source: DTN: MO0403SPAIPCHM.004. 
Figure 3-7. Commercial Spent Nuclear Fuel Ionic Strength Profile with Error Bars 
NOTE: C22C25 represents commercial SNF, 1.5-L/yr water flux, calcium–pore water, 10% fuel exposure, at 25°C. 
July 2004 
3.5.3 Abstraction of Carbonate and Fluoride Concentrations 
Total carbonate is used in the kinetic rate expression for the dissolution of commercial SNF; 
therefore, abstracted values are needed for the TSPA-LA model. In a system where the fugacity 
3-22 No. 7: In-Package Environment Revision 1 
of carbon dioxide (fCO2) is constant over the modeled period and the pH and temperature are 
known, the total carbonate can be calculated using the equilibrium mass action expressions: 
Total C = fCO (Eq. 3-16) 2 (10K 
1 + 10(pH + K 
1 
+ K 
2
) + 10(2pH + K 
1 
+ K 
2 
+ K 
3
)) 
where 
K1 = log K for CO2(g) . CO2(aq); 
K2 = log K for CO2(aq) . H+ + HCO3 
.; and 
K3 = log K for HCO3 
. . H+ + CO3 
2.. 
The values of log K for the above reactions are shown in Table 3-10. 
Table 3-10. Log K Temperature Interpolation Functions for Use in the Total Carbonate Abstraction 
Log K Log K expressiona 
Log K 1 = 7 �~ 10.5T2 . 0.0159T . 1.1023 K 1 
K 2 Log K 2 = 5 �~ 10.7T3 . 0.0002T2 + 0.0132T . 6.5804 
R2 
0.9992 
1.0 
0.9977 K 3 Log K 3 = .8 �~ 10.5T2 + 0.0128T . 10.618 
Source: BSC 2004b, Table 86. DTN: MO0404SPAIPCHM.005. 
NOTE: aT = temperature in degrees Celsius 
The effect of fluoride concentration in the in-package solution is not a direct input parameter in 
TSPA-LA. However, actinide element solubilities are sensitive to fluoride concentrations so the 
ranges of fluoride concentrations were necessary in developing the dissolved concentration 
model (Section 6.2, below). Commercial SNF does not contain fluoride; therefore, the sole 
source of fluoride in commercial SNF waste packages is from the initial seepage waters, which is 
typically low. In contrast, the HLW contains a trace amount of fluoride, which will contribute 
the dissolved fluoride concentration. Based on the EQ3/6 simulations, the upper limit of fluoride 
concentration for codisposal waste packages can be as high as 0.011 mol/kg. 
3.6 IMPLEMENTATION FOR IGNEOUS INTRUSION SCENARIO: 
POSTINTRUSIVE CHEMICAL ENVIRONMENT 
For an igneous intrusion scenario, waste packages in a drift intersected by intrusion will be 
completely damaged and encapsulated by basaltic magma, which will cool and solidify, forming 
a fractured basaltic rock. It is assumed that the postintrusive chemical environment will be 
dominated by the interaction of seepage water percolating through the rock fractures and the 
basaltic rock matrix. In this sense, the chemical environment discussed in this section is no 
longer in-package chemistry per se. 
The chemical interaction of seepage with the basalt is simulated with EQ3/6 for typical seepage 
water and basaltic rock compositions and flow rates (BSC 2004d, Section 6.5.3). The simulation 
results are summarized in Table 3-11. 
July 2004 3-23 No. 7: In-Package Environment Table 3-11. Value of pH and Ionic Strength for Use in the Igneous Intrusion Scenario 
log(fCO2) 
.2 bar 
.2.5 bar 
.3 bar 
.3.5 bar 
Time Period* 
x ¡Ü 25 years 
25 > x ¡Ü 250 years 
250 > x ¡Ü 2500 years 
2500 > x ¡Ü 20000 years 
x ¡Ü 25 years 
25 > x ¡Ü 250 years 
250 > x ¡Ü 2500 years 
2500 > x ¡Ü 20000 years 
x ¡Ü 25 years 
25 > x ¡Ü 250 years 
250 > x ¡Ü 2500 years 
2500 > x ¡Ü 20000 years 
x ¡Ü 25 years 
25 > x ¡Ü 250 years 
250 > x ¡Ü 2500 years 
2500 > x ¡Ü 20000 years 
x ¡Ü 25 years 
25 > x ¡Ü 250 years 
250 > x ¡Ü 2500 years 
2500 > x ¡Ü 20000 years 
.4 bar 
Source: BSC 2004d, Table 8-2. 
NOTE: Time period represents time after re-establishment of seepage flow. Values archived in output 
DTN: MO0402SPAHWCIG.002. 
Maximum and minimum pHs were taken from individual EQ3/6 runs. It can be seen from Table 
3-11 that the pH of the reacted seepage waters will be maintained at a neutral to mildly basic 
range and the ionic strength will remain less than 0.2 mol/kg. EQ3/6 simulations also show 
either no changes or decrease in dissolved fluoride concentration during seepage¨Cbasalt 
interactions. Therefore, for the TSPA implementation, the fluoride concentration, if needed, 
should be set to the initial seepage water concentration. 
3.7 UNCERTAINTIES, LIMITATIONS, AND MODEL CONFIDENCE 
When water comes into contact with the materials inside a waste form, the reactions that ensue 
will be the same as, or very similar to, ones that occur naturally in soils and in other man-made 
materials. Reduced metals, metal oxides, and glass will dissolve and secondary corrosion 
products will form, all in the presence of ambient oxygen and carbon dioxide. In soils, the 
balance between proton production and consumption maintains the pHs to roughly neutral 
values. Only in highly unlikely scenarios do soil pHs fall below 4 (in essence, when high levels 
of reduced sulfur are present). Soil pHs exceed 8 to 9 only when the soil fluids are out of contact 
with the atmosphere and low carbon dioxide fugacities prevail. The in-package chemistry model 
faithfully reproduces these trends and limits. 
3-24 No. 7: In-Package Environment 
pH 
Minimum 
7.6 
7.8 
7.8 
7.7 
8.1 
8.3 
8.3 
8.2 
8.6 
8.8 
8.8 
8.7 
9.0 
9.2 
9.2 
9.1 
9.4 
9.6 
9.6 
9.6 
Maximum 
7.9 
8.1 
8.1 
8.1 
8.4 
8.6 
8.6 
8.6 
8.9 
9.0 
9.1 
9.1 
9.4 
9.5 
9.6 
9.5 
9.7 
9.9 
9.9 
9.9 
Revision 1 
Ionic Strength (mol/kg) 
Maximum 
1.87 ¡Á 10.2 
3.59 ¡Á 10.2 
4.00 ¡Á 10.2 
3.58 ¡Á 10.2 
2.09 ¡Á 10.2 
4.33 ¡Á 10.2 
5.00 ¡Á 10.2 
4.22 ¡Á 10.2 
2.43 ¡Á 10.2 
5.00 ¡Á 10.2 
5.00 ¡Á 10.2 
5.00 ¡Á 10.2 
3.00 ¡Á 10.2 
5.00 ¡Á 10.2 
5.00 ¡Á 10.2 
5.00 ¡Á 10.2 
3.42 ¡Á 10.2 
5.00 ¡Á 10.2 
5.00 ¡Á 10.2 
5.00 ¡Á 10.2 
Minimum 
9.64 ¡Á 10.3 
1.23 ¡Á 10.2 
1.22 ¡Á 10.2 
1.15 ¡Á 10.2 
9.23 ¡Á 10.3 
1.31 ¡Á 10.2 
1.29 ¡Á 10.2 
1.18 ¡Á 10.2 
9.14 ¡Á 10.3 
1.40 ¡Á 10.2 
1.37 ¡Á 10.2 
1.19 ¡Á 10.2 
9.44 ¡Á 10.3 
1.67 ¡Á 10.2 
1.62 ¡Á 10.2 
1.30 ¡Á 10.2 
1.02 ¡Á 10.3 
1.97 ¡Á 10.2 
1.89 ¡Á 10.2 
1.43 ¡Á 10.2 
July 2004 Revision 1 
The in-package chemistry model combines two approaches to model uncertainty: (1) application 
of a factorial design approach to account for known large potential variations in model input 
(reactant combinations, water flux, fuel exposure, temperature, and seepage composition); and 
(2) sensitivity analysis of lesser known, not as well defined input variations (carbon dioxide 
fugacity, Carbon Steel Type A516 sulfur content, corrosion rates, extreme temperatures and flux 
values). Where the first approach provided the functional basis of the in-package chemistry 
model abstraction, the second approach provided the uncertainty ranges of the abstracted 
parameters. Thus, the information to be used in TSPA-LA directly incorporates uncertainty 
exterior to the in-package environment and interior to the in-package environment in a form that 
can be readily implemented in TSPA-LA (i.e., model uncertainty is propagated through the 
abstractions). Thus, the only restrictions on the subsequent use of the in-package chemistry 
model abstraction in TSPA-LA are that all of the abstractions must be applied within the stated 
limits (flux, temperature, fCO2, and fuel exposure) as specified in In-Package Chemistry 
Abstraction (BSC 2004b). 
The in-package chemistry model was validated in two parts. The general output of EQ3/6 was 
validated by comparison with field observations (BSC 2004b, Section 7). Specifically, the 
buffering capacity of the in-package environment was considered quantitatively and shown to be 
reasonable relative to similar natural cases. The reactions that cause long-term shifts from 
neutral pH in natural systems (e.g., acid generation from metal sulfide oxidation, alkalinity 
production by base silicate dissolution in the absence of atmospheric carbon dioxide) were 
shown to be absent in the in-package environment. The modeling of corrosion product surface 
chemistry was validated through expert review, which concurs that the surface effect approach is 
reasonable (BSC 2004b, Section 7). 
3.8 SUMMARY 
Based on the modeling results presented in In-Package Chemistry Abstraction (BSC 2004b), 
in-package chemistry is insensitive to initial water composition, fCO2, decreased corrosion rate 
of the waste package alloys, and minor modifications to waste package design configuration. 
The major controlling factors include waste package type (commercial SNF versus codisposal 
waste package), water flux, fuel exposure (cladding failure) and temperature. The EQ3/6 
calculations establish the range of in-package chemical compositions as dictated by dissolution 
and reprecipitation of solids. Effluent chemistries will be strongly affected by the presence of 
the surfaces of corrosion products (primarily iron oxides and hydroxides). Corrected for the 
effect of surface complexation on corrosion products, the in-package pH ranges from 4.5 to 8.1 
for the nominal and seismic scenario classes. For the igneous intrusion scenario class, the pH is 
neutral to slightly basic (7.6 to 9.9). The ionic strength inside a waste package increases with 
time and can reach a value above 1 mol/L. Increasing temperature and carbon dioxide levels 
would favor decreased pH and vice versa. Higher temperatures cause a general shift in acid-base 
equilibria towards lower pHs. Carbon dioxide levels greater than ambient lower pH as well 
(both inside the waste form and in a basalt-dominated fluid). 
July 2004 3-25 No. 7: In-Package Environment INTENTIONALLY LEFT BLANK 
3-26 No. 7: In-Package Environment 
Revision 1 
July 2004 Revision 1 
4. COMMERCIAL SPENT NUCLEAR FUEL CLADDING DEGRADATION 
Most commercial SNF is encased in Zircaloy cladding. Therefore, degradation of commercial 
SNF considers degradation of the Zircaloy cladding, which can fail either before or after 
emplacement at the repository. The cladding degradation model is implemented in all three 
scenario classes, although in the igneous intrusion modeling case of the igneous scenario, the 
cladding in the impacted drifts is assumed to provide no protection for the fuel. Hence, the only 
aspect of the cladding degradation model implemented in the igneous intrusion modeling case is 
calculating the surface area available for radionuclide diffusion from the waste form to the 
corrosion products in the waste package. The fuel rod cladding contained in drifts that are not 
affected by an igneous intrusion event are treated as those in the nominal scenario. A small 
fraction (1.04%) of fuel rods is encased in stainless steel cladding rather than Zircaloy cladding. 
The cladding model for these fuel rods is bounding in that the cladding is assumed to split along 
the entire length of the fuel rod when the waste package fails. 
The relevant processes and assumptions supporting the cladding degradation model are presented 
below, followed by the cladding degradation model abstraction. A discussion of the sources of 
data, uncertainty, and model confidence building activities concludes this section. The primary 
sources of information for this section are Clad Degradation–Summary and Abstraction for LA 
(BSC 2003b), Clad Degradation–FEPs Screening Arguments (BSC 2004a), and Seismic 
Consequence Abstraction (BSC 2003c). 
July 2004 
4.1 RELEVANT PROCESSES AND ASSUMPTIONS 
Zircaloy cladding degradation due to general corrosion, localized corrosion (radiolysis-enhanced, 
pitting, crevice, and fluoride-enhanced), the presence of dissolved silica, creep rupture, internal 
pressurization, mechanical impact, diffusion-controlled cavity growth, rockfall, electrochemical 
effects, chemical effects, thermal expansion or stress of in-package components, stress corrosion 
cracking, and hydride embrittlement is not expected for repository conditions (BSC 2004a, 
Table 1-1). Some Zircaloy cladding will degrade before disposal, and those fuel rods that have 
degraded cladding will split when contacted by water as a result of the increased volume of 
corrosion products. Because naval SNF is considered to be commercial SNF for modeling 
purposes, naval SNF cladding is assumed to be affected by the same processes (BSC 2004a, 
Table 1-1). 
The cladding degradation model consists of two phases: cladding failure (perforation) and the 
subsequent splitting of the cladding along the length of the fuel rod. The first phase, cladding 
perforation, is the formation of small cracks or holes in the cladding from various sources. In the 
absence of the cladding degradation mechanisms given above that are not expected for repository 
conditions, cladding perforations can occur (BSC 2003b, Section 5.2): 
1. During reactor operation and subsequent storage (including other operations before 
being received at the repository) 
2. In the repository by mechanical action resulting from seismic events that can occur at 
any time (i.e., the seismic scenario class) 
4-1 No. 7: In-Package Environment Revision 1 
3. In the repository from static loading of rock overburden once the drip shield and waste 
package have failed after the 10,000-year regulatory period. 
In the nominal scenario, only the small fraction of Zircaloy cladding with preexisting defects 
(cause 1, above) is assumed to be perforated during the 10,000-year regulatory period. All 
stainless steel cladding is assumed to be perforated when the waste package fails. Perforation of 
cladding from seismic events (cause 2, above) is modeled exclusively in the seismic scenario 
class. Perforation of cladding from rockfall (cause 3, above) is part of the nominal scenario class 
but occurs after the 10,000-year regulatory period. 
Once the cladding has been perforated and the waste package containing the perforated cladding 
has failed, the cladding is assumed to immediately split along the length of the fuel rod, exposing 
the fuel to corrosion. This is the second phase of cladding degradation and is shown in 
Figure 4-1. Corrosion of the fuel occurs primarily along the cladding-fuel interface and along 
fractures within the fuel matrix. In this phase, corrosion of the fuel forms secondary mineral 
phases composed of higher oxides of UO2 (commonly called rind), causing the cladding split to 
widen further and creating a diffusion pathway for radionuclides and water. Perforation and 
splitting of cladding affects the diffusion of radionuclides from the rind into the corrosion 
products in the waste package. 
Figure 4-1. Fuel Rod Showing Cladding, Fractured Fuel, and Cladding Split 
4-2 July 2004 No. 7: In-Package Environment Revision 1 
A small percentage (1.04%) of commercial SNF has stainless steel cladding instead of Zircaloy 
cladding. Stainless steel cladding was used in early core designs but is no longer used, so the 
quantity of stainless-steel-clad fuel is known and is unchanging. Stainless-steel-clad fuel rods 
are assumed to be perforated upon emplacement in the repository, making them immediately 
available for splitting lengthwise when the waste package fails (BSC 2003b, Section 6.2.2). It is 
assumed in this model that there is a finite probability that any delivery of fuel rods to the Yucca 
Mountain site containing stainless-steel-clad fuel rods will be loaded into two consecutive waste 
packages. This assumption is necessary because it addresses an absence of direct confirming 
evidence of how the waste packages are to be loaded. The basis of this assumption is that the 
stainless-steel-clad fuel is loaded into waste packages as it is received at the repository facilities 
to simplify surface facility operations. The model addresses the probability that for any 
stainless-steel-clad fuel delivery, the waste package being loaded could be near full and the 
remaining stainless-steel-clad fuel assemblies would then be loaded into a new waste package, 
increasing the number of waste packages containing stainless-steel-clad fuel assemblies. The 
shipments are not large enough to sequentially load three waste packages for a given delivery 
(BSC 2003b, Section 5.1). 
Naval SNF cladding is modeled as commercial SNF cladding. Naval SNF has been found to be 
more robust than commercial SNF. Using a naval SNF source term developed by the Naval 
Nuclear Propulsion Program, the DOE estimated that the dose resulting from naval SNF is about 
four orders of magnitude less than that from an equivalent amount of commercial SNF 
(BSC 2001, Section 6). A planned classified Naval Nuclear Propulsion Program Addendum to 
the LA will provide details of this naval SNF source term. Therefore, using commercial SNF as 
a surrogate for naval SNF will overestimate radionuclide releases from naval SNF. 
4.2 MODEL FOR DEGRADATION OF CLADDING 
The model for degradation of commercial SNF cladding is presented in four parts: 
• Fraction of cladding that is defective prior to placement in the repository 
• Number of waste packages containing stainless steel cladding 
• Mechanical failure of cladding after placement in the repository 
• The area available for diffusion and volume of water available for UO2 corrosion after 
the cladding has split. 
The justification for the cladding degradation model used in TSPA-LA is based in large part on 
the exclusion of a number of processes considered unlikely to occur in the repository or to have 
an insignificant impact on repository performance. Some of the processes excluded from the 
cladding degradation model are general corrosion of cladding, localized corrosion from 
radiolysis, localized corrosion in pits and in crevices, creep rupture of cladding, stress corrosion 
cracking, and fluoride-enhanced localized corrosion of cladding. The basis for excluding these 
processes is given in Clad Degradation—FEPs Screening Arguments (BSC 2004a, Section 6). 
4-3 July 2004 No. 7: In-Package Environment Revision 1 
4.2.1 Fraction of Cladding Perforated Prior to Placement in the Repository 
The percent of fuel rods with cladding that has perforations prior to placement in the repository 
is given by a loguniform distribution with a range of 0.01% to 1% and a median value of 0.1. 
The loguniform distribution was selected because it gives equal weight to each decade of the full 
distribution. This failure rate is based on an analysis of reactor historical data by S. Cohen & 
Associates (1999, p. 2-31). It also includes defects from wet and dry storage at reactor sites and 
from handling of the waste packages. The as-received perforated fuel rods are available to 
undergo clad splitting and fuel-pellet corrosion once the waste package fails. 
S. Cohen & Associates (1999, p. 2-31) concluded that damage of fuel during shipment, if any, 
will be minor. The rate of cladding perforation resulting from damage during handling at the 
repository surface facility is considered to be similar to that from damage during reactor 
operation and was estimated to be 0.0003% (S. Cohen & Associates 1999, p. 7-1) to 0.0005% 
(CRWMS M&O 2001a, Section 6.4.1). 
(Eq. 4-1) 
4.2.2 Number of Waste Packages Containing Stainless-Steel Cladding 
The fraction of fuel rods that have stainless steel cladding is 1.04% of the total commercial SNF 
inventory. Fuel with stainless steel cladding is placed into waste packages as it arrives at the 
repository. Assuming that deliveries of fuel with stainless steel cladding can be placed into 
either a single waste package or two consecutive waste packages results in 3.5% to 7% 
(uniformly distributed) of the waste packages containing stainless-steel-clad fuel rods (see 
Section 4.3). The percentage of stainless-steel-clad rods in a waste package is calculated in 
Equation 4-1 as: 
% SS - clad rods in waste package = (100% × 1.04%) 
clad fuel rods containing SS - % of waste packages 
Thus, waste packages containing stainless-steel-clad fuel rods will contain 15% to 30% 
stainless-steel-clad fuel rods, which are modeled as perforated and available for instantaneous 
splitting when the waste package fails. 
4.2.3 Mechanical Failure of Cladding after Placement in the Repository 
Once the fuel has been emplaced in the repository, intact cladding can be perforated as a result of 
seismic events that are modeled exclusively in the seismic scenario class (vibratory ground 
motion, fault displacement, and rockfall induced by ground motion) or as a result of the static 
load of a rock overburden. These two causes of mechanical failure are discussed below. 
Seismic Events–The seismic events representing the seismic scenario class can occur at any 
time, and the effects of seismic activity on cladding integrity are developed and analyzed in 
Seismic Consequence Abstraction (BSC 2003c). Seismic events can damage fuel cladding in 
two ways: damage to the cladding from end-to-end impacts between adjacent waste packages 
and damage to the cladding from fault displacement. The end-to-end impact between two 
adjacent waste packages accounts for 87% of the mean damage to the waste package at the 10-6 
per year ground motion level and 92% of the mean damage to the waste package at the 10-7 per 
4-4 July 2004 No. 7: In-Package Environment Revision 1 
year ground motion level (BSC 2003c, Section 6.5.1.2). These results imply that the end-to-end 
impact of adjacent waste packages produces much more severe forces and accelerations than the 
side-on impact between a waste package and the emplacement pallet or drip shield. 
In terms of peak ground velocity, the abstraction for damage to the cladding during a seismic 
event is a simple lookup table with a linear interpolation between the two points in Table 4-1, 
which shows that cladding is assumed to be 100% damaged if the peak ground velocity exceeds 
1.067 m/s. The peak ground velocity values of 0.55 and 1.067 m/s correspond to mean annual 
exceedance frequencies of seismic events of 5 × 10-5 per year and 10-5 per year, respectively. 
There is no uncertainty in this abstraction because it represents a bounding estimate for cladding 
response at all values of peak ground velocity. 
Table 4-1. Abstraction for Damage to the Cladding from Vibratory Ground Motion 
Damage to Cladding 
(%) 
0