Text

Forwards CNWRA 93-006, Characteristics of Spent Nuclear Fuel & Cladding Relevant to HLW Source Term
	ML20056D854
Person / Time
Issue date:	08/06/1993
From:	Joseph Holonich; NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:	Shelor D; ENERGY, DEPT. OF
References
	REF-WM-11 HLWR, NUDOCS 9308180162
	Download: ML20056D854 (1)
	v • d • e

WS 0 61993 0003

,,Mr. Ot5ight E. Shelor, Associate Director for Systems and Gompliance Office of Civilian Radioactive Waste Management U. S. Department of Energy 1000 Independence Ave, S.W.

Washington, D. C.

20585

Dear Mr. Shelor:

Enclosed for your information is the Center for Nuclear Waste Regulatory Analyses (CNWRA) report entitled " Characteristics of Spent Nuclear fuel and Cladding Relevant to High-Level Waste Source Term." The enclosed report was prepared to document work performed by the CNWRA for the Nuclear Regulatory Commission. The Department of Energy may find the information in this report use ful.

If you have any questions, please feel free to contact Charlotte Abrams of my staff at (301) 504-3403.

SincArely, Josd@ J. Holonich, Director Repository Licensing and Quality Assurance Project Directorate Division of High-Level Waste Management Office of Nuclear Material Safety and Safeguards

Enclosure:

As stated cc:

R. Loux, State of Nevada T. J. Hickey, Nevada Legislative Committee C. Gertz, DOE /NV M. Murphy, Nye County, NV M. Baughman, Lincoln County, NV D. Bechtel, Clark County, NV D. Weigel, GA0 P. Niedzielski-Eichner, Nye County. NV B. Mettam, Inyo County, CA V. Poe, Mineral County, NV

f. Sperry, White Pine County, NV R. Williams, Lander County, NV L. Fiorenzi, Eureka County, NV

~

J. Hoffman, Esmeralda County, NV C. Schank, Churchill County, NV L. Bradshaw, Nye County, NV DISTRIBUTION CNWRA NMSS R/F HLPD R/F LSS LPDR ACNW PDR CENTRAL FILE BJYoungblood, HLWM JLinehan, HLWM RBallard, HLGE Mfederline, HLHP OFC HLPD b

HLP)\\/ b HLRD4/

8 NAME AGarcia h JYthms dMdoOch DATE 08/@/93 08/[o/93 08/[i/93 C = COVER E = COVER & ENCLOSURE N = NO COPY S:\\93-006 0FFICIAL RECORD COPY t

I s.b

\\\\

\\

+i^

f 9308180152'930806 PDR WASTE s

WM-13 PDR j

CNWR A 93-006 I

i w,,;*,,q'%=s y;)::w:7virmwwww??w' W~ #"M*W*M"*WM W'W***"W?W"Wi-n

,6 uy??My ( 7,,y%

r?

fLw-e2MQ m :f +::4;>r.e u; w n s.~ y m - m a l.a.u:. w a p K =v &"% Mt#W :

n n;%MM w t

M p M N 7 ;N % S G ""? N & T"T"'" T M N M W " " 9 %.$

.,,.. ~m.* i,m o:M wq,n.

+ :

.; p

,, #,p m,

.m, g..

4 c

I~,--

j..,T

.-}

> NI,.. - gg g

g..g

,. w g.

.; g,, g,z - -

o g

yf jr

,j. $(

- ;. f 3> % g.7,,*

i k

"i

>f' &l 1 m e m m m))'"d 0 u*e*..

A

'd R

: 4 m

. N

'sd5 '.

~*\\' -:NQ ik -

bdfMENN^^f QAda p-t a 94 31 04(y.yac B D M"Q m y W b & g.. Vd % m./p A & g e m nd em 9

A 4

4 J

4

. 4m 3yy u g m%gp w; kw p+Lg/ ked eva% 0

[,

MS4 V

p g9 h

w wx4

-y v

n "L.h e, gg ;WlW w

I"# d ddie faad.ie A M uu;&

N Wh.64 2UNt* "-

' $Xseu.dshnakt.ul -- i I.*P'**TP7(PWMPWTP"'P""EPS'M"I'5PS*'"WNPWT i

{.%,,.

F

'#WY7"fMT,9 I

%@f M

f'Q %;

Q

&a;[]%45ad a

r i 2 -l

<%~

~

~ l~

A & lAz i di.CEin da n k b i kN h M i a A S k % W '

~

l 1 8 l

l l

N l 1 l

,u,,

Nuclear Regulatory Commission Contract NRC-02-88-005 3a

@a It n.

k m w-

& % Regulatory Analyses w.s ';

e,, n,i.n Tn L,, s..

s

-s s vh l

l 1 < i o' } h[

t v,' v v

3 3

R >

~

I.

3 CNWRA 93-006 i

i i l t

i i

Prepared for j!

Nuclear Regulatory Commission Contract NRC-02-88-005 g

l l

)

Prepared by

)

ll Center for Nuclear Waste Regulatory Analyses San Antonio, Texas l

'l May 1993 ll 4

1 l

,-.7

-.[.

-..e-

-a--

- - - =- - - - ~ ' - - - - *---**

CNWRA 93-006 I

l CHARACTERISTICS OF SPENT NUCLEAR FUEL AND CLADDING RELEVANT TO I

HIGH-LEVEL WASTE SOURCE TERM I

I I

Prepared for Nuclear Regulatory Commission Contract NRC-02-88-005 I

I i

Prepared by I

Hersh K. Manaktala I

'g Center for Nuclear Waste Regulatory Analyses San Antonio, Texas I

I I

May 1993 I

l PREVIOUS REPORTS IN SERIES Number Name Date Issued CNWRA 92-017 An Assessment of Borosilicate Glass as a High-Level Waste Form September 1992 CNWRA 92-018 Leaching of Borosilicate Glass Using Draft ASTM Procedure for High-Level Waste August 1992 I

l

. I I

I I

I ii

I ABSTRACT I-This report, based on literature study, describes characteristics of light water reactor (LWR) fuel assemblies for boiling water reactors (BWR) and pressurized water reactors (PWR) and the changes that take place in both cladding and uranium dioxide fuel during service in commercial power reactors. This I

information is provided as a background for the evaluation of important factors related to fuel stability under geologic repository conditions. Data related to discharged fuel storage (both wet and dry) are also provided, along with the condition of the fuel in terms of damaged and leaking fuel assemblies. h degradation of spent fuel and cladding while in service in the reactor and likely degradation in a geologic I

repository are discussed in terms of cladding oxidation and corrosion, and fuel-pellet cracking, fuel restructuring, microstructure and fission product mobility, inventory and distribution of fission products, fuel pellet rim effect, and fission gas release and pressure increase. The range of attributes of discharged I

fuel will have an impact on the releases of radionuclides from the engineered barrier system (EBS) and, as a result, on the compliance with the regulations relating to the gradual release from a repository over a peritz of 10,000 years. A suggestion has been made to characterize a wider variety of spent LWR fuels, including high burnup fuels and fuels with characteristics outside the range of those currently awaiting geologic disposal, with the intent of including the full range of spent fuel characteristics and performance in developing source term models for performance assessment. A significant portion of this report deals with review of studies on oxidade: and release of radionuclides from spent fuels. Although the emphasis is on an oxidizing environment, the likely behavior of spent fuel under a reducing environment has also been discussed. The parameters important for assessment of spent fuel behavior in a geologic repository include oxidation state of the fuel at the time of contact with water, geochemistry of the repository (chemical composition of the leachant), water flow conditions, and synergistic effects due to modification of the geochemical environment as a result of releases from vitrified waste form and I

corrosion products from the waste package. A number of areas requiring additional experimental data have been identified. Issues related to modeling of source term for use in performance assessments have also been discussed.

I I

I iii I

1

i I

CONTENTS Section Page 1

INTRODUCTION 1-1

. I 2

CLADDING, FUEL, AND FUEL ASSEMBLY CHARACTERISTICS (AS-FAB RICATED).....................

2-1 I

2.1 FUEL CLADDING MATERIAL................................. 2-1 2.2 FUEL ROD AND ASSEMBLY DESIGN............................

2-1 3

REACTOR SERVICE INDUCED CHANGES...........

3-1 3.1 DEGRADATION OF CLADDING IN THE REACTOR............

3-1 3.2 DESIGN MODIFICATIONS............

3-1 3.2.1 Boiling Water Reactor Fuels...............

3-1 3.2.2 Pressurized Water Reactor Fuels..........................

3-2 3.3 DEGRADATION OF FUEL IN THE REACTOR..

3-2 3.3.1 Fuel-Pellet Cracking..........

3-5 3.3.2 Fuel Restmeturing 3-5 3.3.3 Microstructure and Fission-Product Mobility.......

3-5 l

l 3.3.4 Inventory and Distribution of Fission Products.........................

3-6 3.3.5 Fuel Pellet Rim Effect 3-6 i

3.3.6 Fission Gas Release and Pressure................................. 3-11 4

SPENT FUEL AND CLADDING CHARACTERISTICS, STORAGE, i

A ND I NVENTORY........................................

4-1 l[

4.1 RADIOACTIVITY ASSOCIATED WITH CLADDING.................

4-1 l5 4.1.1 Radioactivity as a Function of Fuel Burnup..

4-1 4.1.2 Distribution of Radionuclides.........................

4-1 lg 4.2 RADIOACTIVITY IN SPENT FUEL......

4-10

'E 4.3 CHARACTERISTICS OF SPENT FUEL......................

4-11 4.4 STORAGE OF SPENT LWR FUEL.

4-11 4.4.1 U.S. Spent FueI Inventory..........

4-12 I

r 5

POST-DISCHARGE SPENT FUEL DEGRADATION...

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

~

5.1 SPENT FUEL OXIDATION..

5-1 I

5.1.1 Fuel Condition Dependence.......

5-1 5.1.2 Temperature Dependence................

5-3 5.1.3 Spent Fuel Corrosion and Degradation...............

5-5 I

5.1.4 Oxic Conditions..........

5-6 5.1.4.1 Release of Actinides..........

5-6 5.1.4.2 Release of Fission Products....................................

5-6 r

l 5.1.5 Reducing Conditions.......................................

5-9 5.1.6 Comparison of Leaching Under Oxidizing Versus Reducing Environment 5-9 5.1.7 Fuel Burnup Effects

..................................... 5-11 I

iV E

l l

I...-

I CONTENTS (cont'd)

Section Page 5.1.8 Flowthrough Leaching Tests..

5-11 I

5.1.9 Unsaturated Repository Simulating Tests........................

5-16

" " " " " " " " * " " " " " ^ * " ' " " - ~ '

I 7

REFERENCES

...................................7-1 I

1 I

I I

I

FIGURES Figure Page 2-1 Schematic of fuel assembly (a) PWR fuel; (b) BWR fuel..................

2-2 2-2 Highest and average burnup of LWR spent fuel assemblies currently in storage

{

(Bailey and Johnson, 1990)...................................... 2-3 3-1 Fuel rod failures in LWRs for the years 1969-1977 (Garzarolli and Stehle,1979)... 3-3

{

3-2 Fuel rod failure trends in LWRs for the years 1969-1977 (Garzarolli and Stehle, 1979)....................................................

3-4 3-3 Typical porosity near the fuel pellet edge (Burnup: on ATM-103 -

33 mwd kgU-1; on ATM-104 - 44 mwd kgU-1) (Guenther et al.,1990)....

3-11 3-4 Calculated pressure increase caused by helium production from actinide decay (Johnson and Gilbert, 1983)..................................... 3-12 3-5 Comparison of fission gas release from unpressurized and pressurized LWR fuel rods as a function of the fuel burnup (Ocken,1982)..............

3-13 41 Gross activities of a emitters on the cladding surfaces as a function of fuel burnup (Hirabayashi et al., 1991)....................................... 4-3 4-2 Elemental contribution to a activity in the spent nuclear fuel as a function of fuel burnup (Hirabayashi, et al.,1991).............................. 4-3 4-3 Activity of # or y emitters in the spent fuel claddings (Hirabayashi, 1991).......

4-4 4-4 Radionuclide distributions in Zircaloy cladding specimens (Brodda and Merz,1984).

4-9 5-1 O/M ratio with oxidation time in flowing dry air at 195*C (Einziger et al.,1992)..

5-2 5-2a Time rate of change in O/M ratio as a function of oxidation time up to 5500 hours0.0637 days 1.528 hours 0.00909 weeks 0.00209 months (Einziger et al., 1992 )......................................... 5-4 5-2b Time rate of change in O/M ratio as a function of oxidation time up to 3000 hours0.0347 days 0.833 hours 0.00496 weeks 0.00114 months (Einziger et al., 1992)......................................... 5-4 5-3 Schematic of O/M ratio in spent fuel as a function of oxidation time (Einziger et al.,1992)

.................................5-5 5-4 Uranium concentration in deionized water (DIW) and synthetic groundwater (SGW) l as a function of cumulative contact time under oxic conditions (Forsyth and Werme, 1992 ).............................................. 5-7 5-5 Plutonium concentrations in DIW and SGW as a function of cumulative contact time under oxic conditions (Forsyth and Werme,1992)......................

5-7 5-6 Fractional release rates for 137Cs under oxic conditions (adapted from Forsyth and Werme,1992) 5-8 5-7 Fractional release rates for "Sr under oxic conditions (adapted from Forsyth and Werme,1992) 5-8 L

5-8 Fractional release rates for "Tc under oxic conditions (Forsyth and Werme, 1992)..

5-9 5-9 Comparison between fractional releases of uranium, plutonium, and fission products under oxidizing and reducing conditions in a low burnup BWR fuel (adapted from Werme and Forsyth,1988)............................ 5-12 vi

_a

B ricuass (cent a>

i l

Figure Page 1

5-10 Comparison between fractional releases of uranium, plutonium, and fission products under oxidizing and reducing conditions in a PWR fuel (adapted from Werme l

and Forsyth,1988) 5-12 137

!g 5-11 Cumulative release fractions for Cs under oxic conditions for 7,28,91,182, 3

and 364-day leaching periods (Forsyth,1991)...............

5-13 M

l 5-12 Cumulative release fractions for Sr under oxic conditions for 7, 28, 91,182, and 36klay leaching periods (Forsyth,1991)..........

5-13 I

5-13 Dissolution rate of radionuclides from spent fuel grains in 2 x 10-2M NaHCO /Na:CO solution at 20 to 25"C and pH 8.0 to 8.2 (Gray et al.,1992).... 5-15 j

3 3

5-14 Dependence of U concentration on reciprocal flow for spent fuel grains in 2 x 10-2M NaHCO /Na2CO solution at 20 to 25*C and pH 8.0 to 8.2 (Gray 3

3

}

et al.,1992)

......................................... 5-15 L

5-15 Interpretive paragenetic sequence formed on top surfaces of altered uraninite pellets.

Onset of U dissolution and pulsed-release periods determined from solution data.

r (Wronkiewicz et al.,1992).........

5-18 1

1 l

l l

l L

vii

5 TABLES Table Page 3-1 Fractional distribution of radionuclides in LWR spent fuel (Wuertz and Ellinger,1985) 3-7 3-2 Activity of selected radionuclides in a PWR fuel assembly irradiated to an average burnup of 33 mwd kgU-1 (adapted from Woodley,1983) 3-8 3-3 Results of sensitivity test cases for 10-year-old fuel (Welch et al.,1990).

3-9 4-1 Concentration of radionuclides in the spent fuel claddings (Hirabayashi et al.,1990) 4-2 4-2 Distribution of the main # and y emitters in spent fuel cledding with 29.4 mwd kgU--I burnup and 5 years cooling (Hirabayashi et al.,1991)...

4-5 4-3 Distribution of the main cr emitters in spent fuel cladding with 29.4 mwd kgU-l burnup and 5 years cooling (Hirabayashi et al.,1991)......

4-6 4-4 Characteristics of cladding specimens investigated (adapted from Brodda and Merz,1984) 4-7 4-5 Percent release of y-emitting nuclides fiom Zircaloy cladding specimens in different teachants (Brodda and Merz,1984)..

4-8 4-6 Percent release of actinides from Zircaloy cladding specimens in different leachants (Brodda and Merz,1984).

4-9 I

4-7 Radioactivity from principal fission products in irradiated fuel (Tsoulfanidis and Cochran,1991).......

4-12 4-8 Radioactivity from principal actinides in irradiated fuel (Tsoulfanidis and Cochran,1991)........

4-13 4-9 Lifetime radioactive waste generation from a PWR and a BWR (adapted from Tsoulfanidis and Cochran,1991).

4-14 g

4-10 Spent fuel discharged during 1968-1990 from com:r.ercial LWRs in the United W

States (U.S. DOE,1992b)...

4-15 4-11 Spent fuel burnup and annual discharged quantities (adapted from U.S. DOE,1992b) 4-16 4-12 Defective BWR fuels by assembly class and fuel design (Moore et al.,1990) 4-18 4-13 Defective PWR fuels by assembly class and fuel design (Moore et al.,1990).

4-19 5-1 Fuel specimen characteristics (adapted from Einziger et al.,1991) 5-2 I

5-2 Ranking of fuels by weight gain (adapted from Einziger et al.,1992).

5-3 5-3 Composition (mmol L-l) of SGW at pH of 8.0-8.2 (Forsyth and Werme,1992).

5-5 5-4 Cumulative release fractions in a 1-y test under oxic and anoxic conditions (adapted I

from Forsyth and Werme,1992) 5-10 55 Rates of fractional release of 337Cs under oxic and anoxic conditions (Forsyth and Werme,1992) 5-10 54 Rates of fractional release of MSr under oxic and anoxic conditions (Forsyth and Werme,1992) 5-10 5-7 Summary of spent fuel dissolution results (adapied from Gray et al.,1992).

. 5-14 5-8 Chemical composition (mg L-1) of EJ-13 water including carbon-containing specien, anions, and cations (pH = 8.2, Solution U = 0.0024) (adapted from Wrontiewicz et al.,1992) 5-16 1

viii lI 1

ACKNOWLEDGMENTS 5

This rcrort was prepared to document work performed by the Center for Nuclear Waste Regulatory 3

Analyses (CNWRA) for the U.S. Nuclear Regulatory Commission (NRC) under Contract No.

NRC-02-88-005. The activities reported here were performed on behalf of the NRC Office of Nuclear Material Safety and Safeguards (NMSS), Division of High-Level Waste Management (DHLWM). The I

report is an independent product of the CSM and does not necessarily reflect the views or regulatory position of the NRC.

I I

E I

I

'I I

I I

I 1

5 I

I

E...-

1 INTRODUCTION The current Code of Federal Regulations, Title 10 Part 60 Section 113, requires the engineered barrier

.I system (EBS) to provide contamment for the high-level radioactive wastes (HLW) for a minimum period of 300 to 1,000 years after permanent disposal in a geologic repository (U.S. NRC,1990). The regulations also specify performance requirements for the EBS following permanent closure of the

,I repository. The maximum allowable release rates of significant radionuclides from the EBS, following the containment period, is specified as I pan in 100,000 per year of the inventory of a particular y

radionuclide calculated to be present at 1,000 years following permanent closure of the repository.

J According to the regulations, the performance of the repository and engineered barriers are to be calculated for a 10,000-year period. However, it should be noted that this is no; intended to be the design life of the repository. Quantitative estimates of near-field releases are also needed to evaluate the far-field migration of radionuclides from the EBS. Such estimates of long-term releases require understanding of property changes of the fuel and modeling of their consequences.

As a step towards developing a better understanding of the important characteristics of the light water reactor (LWR) fuels and cladding that would control the release of radionuclides from the spent fuel in a repository, this report, based on literature, provides an overview of the literature in the field. A number of original figures and tables have been slightly modified or changed. They are shown as ' adapted' from the original reference. All tables and figures used from the literature have been credited to the literature citations. The properties / materials characteristics discussed include composition and radionuclide I

inveraory, fuel-pellet cracking and restructuring, fission gas quantity and compodtion, helium generation from actinide decay, and likely changes in the oxidation state of the failed / defected spent fuel rods and upon breach of the fuel cladding intact at the time of emplacement.

The report also describes and discusses the experimental data reported in the literature on release of radionuclides from unirradiated and spent fuel when exposed to water vapors and aqueous environments.

Processes which may have significant impact on the release of radioactivity from the spent fuel in an unsaturated geologic repository, and which have not been currently addressed in source-term models, have been identified. Incorporation of the suggested processes in the source-term models can be expected to provide more accurate estimates of releases of radionuclides under repository-representative conditions.

I.

The processes identified in the report include: (i) spent fuel and cladding degradation during an extended period of high-tempersture exposura prior to water intrusion, (ii) extended exposure to high-moisture content atmosphere /small amount of water prior to interaction with groundwaters, (iii) release kinetics I

of spent fuel with a higher oxidation state than UO discharged from the reactor / storage, (iv) accounting 3

for the estimated 4,000 to 5,000 failed fuel rods (at the time of the beginning of repository operation) in the source term, (v) modification of groundwater chemistry due to radiolysis of repository air / moisture, I

corrosion products released as a result of waste package degradation / failure, and leached species from I

vitrified wasteform, (vi) additional releases of 'C from the cladding and spent fuel due to oxidation in the repository prior to contacting groundwaters, (vii) fuel cladding failure as a function of time (which I

may be a more conservative approach than the so-called *no-credit

for cladding approach used by the current models), (viii) precipitate and colloid formation phenomena, (ix) microbial activity in the release of radionuclides from spent fuel, and (x) acceleration in the cladding / spent-fuel degradation as a result of galvanic interaction between the cladding, spent fuel, and waste package component materials.

I l-1

?

E.-

~

I 2 CLADDING, FUEL, AND FUEL ASSEMBLY CHARACTERISTICS (AS-FABRICATED) 2.1 FUEL CLADDING MATERIAL 4

Early nucicar power reactors used stainless steel as a fuel cladding material. The search for an i

alternative material with improved performance and lower neutron absorption led to the change from stainless steel to zirconium-based alloys. Only 5 out of approximately 114 pov er reactors in the United I

States (U.S.) used stainless steel cladding (U.S. DOE,1992b). De two commonly used cladding materials for fuels are Zircaloy-2 and Zircaloy-4. Both are dilute alloys with more than 97 weight percent zirconium. Their major alloying elements are tin, iron, chromium, oxygen, and silicon, with

,E Zircaioy-2 also containing nickei. Fuei ciadding for pressurized water reactors (PwR) is primarily

,E fabricated from Zircaloy-4, while Zircaloy-2 is the material for boiling water reactor (BWR) application.

Nickel was removed in Zircaloy-4 to reduce hydrogen pick-up during reactor operation.

l 2.2 FUEL ROD AND ASSEMBLY DESIGN 4

Ig Typical LWR fuel consists of stacks of sintered UO pellets in a Zircaloy cladding. The 2

B cladding is a long cylindrical seamless tube with approximate dimensions of 4 m length,9 to 13 mm q

1 internal diameter, and 0.5 to 0.8 mm wall thickness. The dimensions depend upon the fuel design. The 35

g sealed fuel rods contain UO fuel enriched in U in the form of pellets which are approximately I cm 2

5 m length and slightly smaller in diameter than the fuel cladding internal diameter. A plenum, with a coil J

spring, is provided at the top of the fuel pellet stack for fission gases (Kr, Xe, I ). The rods are arranged 2

l g in a square array to form a fuel assembly. The number of rods in each assembly can vary from 49 to

,g 289 in a 7 x7 to 17 x17 configuration, depending upon the reactor type and the fuel assembly design.

There are at leas. 88 different LWR fuel assembly types which are or have been used in the U.S. (U.S.

l DOE,1992b), and additional ones are evolving (Gilbert et al.,1990). Typical BWR and PWR fuel j

5 assemblies are shown in Figure 2-1. The BWR fuel assembly typically has a width of 14 cra, a length of 448 cm, and a U content of 183 to 195 kg, while the PWR assembly has a width of 47 cm, a length i

of 406 cm, and a U content of 425 to 460 kg (U.S. DOE,1992b). The enrichment of UO, in the fissile isotope of uranium,35U, is approximately 3 percent prior to irradiation, and the fuelhellets have a sintered density of 94 to 95 percent of the theoretical density. Grain size is uniform and is typically 2 to 4 pm. Fuel rods are pressurized with He to improve heat conduction across the fuel-cladding gap, which also tends to reduce fission-gas release. Additional details of the design features of the existing spent fuel inventory are published in the literature (Moore et al.,1990; Moore and Notz,1991).

I The fuel is typically irradiated at an average linear power of 15 to 25 kW m-1 with a fuel pellet centerline temperature in the range of 800*C to 1200*C. Only a small pe*centage of the fuel

(< 10 percent) in the reactor core has a centerline temperature exceeding 1200*C (Woodley,1983). Fuel I

pellet centerline temperature above this limit has a considerable effect on the fission-gas release and microstructural characteristics of the spent fuel. The other major irradiation parameter affecting spent fuel characteristics is the burnup, which is expressed in units of megawatt days per kilogram of uranium I

(mwd kgU-1). Burnups for fuels have typically ranged from 4 to 35 mwd kgU-1 over the past three decades. Data related to the average burnup and highest burnup, separately for BWR and PWR fuel l

assemblies discharged during the years 1968 through 1988, are provided in Figure 2-2. The trend clearly I

burnup of 60 mwd kgU-l and higher.

indicates increasing burnup for both types of assemblies. Presently, PWR fuels are being developed for 2-1 I

1 E..-

I B

i

- = ~~

p n

n E

NOZZLE ASSEMBLY y REMOVABLE TOP D

g H

J y TOP END N:

[

E4 gg g

SPACER GRID

.y-k

p_g UPPER FUEL ROD TIE PLATE -

~

k' FUEL ~

~

CLADDING l

[

y GRID N.,

j l

a

'E

-~

ROD m-_-_

~

SPACER GRID N l

l l

1, I

l REMOVABLE l

y

_ lL., P' 1:

1 c-I f,

- DASHPOT REGON LOWER j:f'.I E

~

?

l TIE PLATE

'*o u

L l

l f

i !

5 N.

h END RTTING I

I Figure 2-1. Schematic of fuel assembly (a) PWR fuel; (b) BWR fuel g

22 I

u

5..-

I 60 PWR f

O # Discharged Fuel with I

Highest Average Bumup

(

8, Maximum

\\i,e s

O E Weighted Average Bumup g

{

50 e

, s sl en of Discharged Fuel l

', /

s I

T l

u

,9-4 y

40 2

A BWR I

Maximum

,e - e' d

p -e' T

,N-E-E-#-M

'E

E'

[c 30 e

g g

I a

to c)

'E O

e

's PWR Weign:ed I

Average 10 BWR Weighted I

Average 5

0 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 Year Figure 2-2. IIighest and average burnup of LWR spent fuel assemblies currently in storage (Bailey and Johnson,1990)

I l

The UO fuels have demonstrated satisfactory dimensional and radiation stability as well as 2

chemical compatibility with the cladding and coolant under both BWR and PWR operating conditions.

However, a number of changes take place in the physical, structural, and chemical charac; eristics of the i

cladding and fuel that could influence radionuclide release under repository environments. These are discussed briefly in this repon.

I I

5..-

3 REACTOR SERVICE INDUCED CHANGES g

3.I DEGRADATION OF CLADDING IN TIIE REACTOR The majority of the past failures of Zircaloy fuel cladding have been attributed to the following sesen modes:

accelerated waterside corrosion e internal hydriding I

pellet-clad-interaction (PCI) e fretting and wear e fuel rod and assembly bow I

e neutron irradiation induced growth a fuel rod collapse or flattening.

Detailed discussion of the above-listed failure modes has been reported in the literature (Garzarolli,1979). The first three modes, which degrade the cladding in the reactor (whether the fuel rod fails or not), are also considered potential degradation modes that are pertinent to geological disposal of spent fuel. The last four degradation modes are primarily applicable to service in the reactor, and are not likely to be operative in a repository environment. However, there are two additional potential degradation modes of interest for the geological disposal of spent fuel; namely, mechanical stress overloading and stress / creep rupture (Manaktala, 1990; 1991).

3.2 DESIGN MODIFICATIONS I

The LWR fuel cladding, fuel rods, and assembly design have evolved over the past two to three decades, resulting in much improvement in the performance-as expressed by the reduction of cladding I

failures in core (Baily et al.,1979). Significant changes that led to the improvement are summarized in the following sections.

3.2.1 Boiling Water Reactor Fuels Fuel rod failures due to Zircaloy hydriding, first observed in BWRs in the early 1970s, have I

been practically eliminated by reducing the amount of moisture introduced during fabrication and by the inclusion of a ' hydrogen getter' within the rod. Moisture was reduced mamly by elevated temperature vacuum drying of the loaded fuel rods. A negligible number, <0.002 percent, of waterside corrosion I

failures was observed in some early cores in BWRs. These failures were attributed to a drastic increase in the coolant pressure drop, due to heavy deposition of copper-rich ' crud' at the fuel assembly inlet nozzles, leading to the blockage of heat flow by corrosion layers or crud deposits. The source of the I

crud products was traced to the use of Cu-Ni alloy as tubing material in feedwater heaters. The problem was eliminated by replacing them with stainless steel fabricated units and by effective demineralization and optimization of the oxygen content of the core coolant. The failure due to PCI, initially observed g

in 1971, has been reduced by: (i) change of reactor operating procedures to avoid sharp ramps in power B

levels, especially after mid-life of the fuel in core, (ii) introduction of fuel rods with smaller diameters, resulting in reduced linear heat output of the rod, (iii) fuel-design improvements, for example, short I

chamfered pellets and annealed cladding, (iv) addition of plasticizer in the fuel pellets, for example, alumino-silicates to reduce the stress level on the cladding upon localized pellet-clad interaction, and 3-1 I

i E..-

(v) use of a low-oxygen unalloyed zirconium liner bonded to the cladding inner surface-that is, a ductile barrier between the fuel and the Zircaloy cladding. Another design change, implemented in 1977, was

.g the increase in helium internal pressure of the fuel rod from 1 to 3 atm. This resulted in reducing the

g fuel operating temperatures. At lower fuel temperatures, fission gas and the more volatile fission products have lower mobility for migration into the fuel-cladding gap through grain boundaries and pellet

a cracks. Reduced migration led to reduced potential for environment embrittlement of the cladding and

g fewer stress corrosion related failures. It is generally agreed that 12 and probably Cs/Cd are the predominant embrittling agents.

i 3.2.2 Pressurized Water Reactor Fuels I

i

'Ihe evolutionary changes in PWR fuel design include the use of greater fuel rod pressurization l

i than in BWR fuel rods to prevent cladding collapse onto the fuel due to the highe: system pressure (fuel rod internal pressure was increased from 1 to about 30 atm of helium) (Andrews et al.,1979). Also, the UO fuel Pellet fabrication parameters were modified to yield pore size density and distribution which 2

j made the fuel less prone to radiation-induced sintering in the reactor. Densification of fuel pellets in service has been known to cause fuel column shortening, formation of gaps in the fuel column, and increase in the gap between the fuel and cladding. This phenomenon has led to cladding collapse in PWRs and, in some cases, to the release of fission products in the coolant. Although fuel rod failures due to PCI have been fewer in PWRs than in BWRs, some of the remedies implemented for BWRs have been introduced in PWR fuel rods or are being considered. These include shorter and chamfered fuel I.

pellets, revised operating guidelines for core power increases (less severe ramp-ups), and use of a zirconium liner.

The frequency of cladding failures has been reduced progressively over the past few decades from 2 percent or more in some BWR cores to 0.01 to 0.02 percent in both BWRs and PWRs, at present.

According to industry estimates, this low failure level is expected to be maintained (Locke,1975; Garzarolli et al.,1979). The frequencies of failures due to various modes reported in the literature are I-illustrated in Figure 3-1. The progressively reducing trends are evident from Figure 3-2. The dramatic improvement in the fuel rod failures, to a large extent, is a result of a series of design modifications, the i

salient ones described earlier. Neither technical nor economical justification is suggested for reducing the fuel failure level below 0.01 percent (Locke,1975). Consequently, in the inventory of spent fuel, one can anticipate at least 1 in every 10,000 rods to be breached. The spent fuel rods with breached cladding obviously pose a much greater risk for release of fission products to the geological repository.

i I

Therefore, additional contribution to the near-field source term needs to be incorporated in performance 1

assessment models/ codes to account for breached cladding. The models should also account for rods that i

may be breached in handling during waste package loading, assembling, and transportation to the repository for emplacement.

3.3 DEGRADA"'ON OF FUEL IN THE REACTOR l

I After irradiate in a nuclear reactor, the composition of the fuel becomes more complex.

Along with the original uranium dioxide, the spent fuel pellets contain gaseous and solid fission products.

1 I

These are generally distrioutM inhomogeneously. Additional variability in the characteristics of the spent fuel results from reactor ope og power levels, location of the fuel in core, and total burnup of the fuel assembly. Some of the significant changes are described as follows.

I 3-2 I

1

10

E Eav ' P o' o ' " '"'-

3 Failure Mode I

(W Plants Affected) 1l Hydriding/PCI i

j 3

(25 Plants With Old i

3 6x6 /7x7 First Core I

(5 pec )

.a u.

0.1 l-N e

~

i N

=

s

.:E I

Unsystomatic Failures (All Plants)

'N' i3 0.01 (Some Hydriding. Cause Frequently Unknown)

)

BWR i

I 0.001 l

1969 1970 1971 1972 1973 1974 1975 1976 1977 Year I

10 1

Envelope Of Data Points Failure Mode i

l i

Clad

(# Plants Affected)

~

Collapsing

-g 1

(5)

l g

PCI(Ramp incidents) 7

I

\\

(3 + 1 Suspect) 2*

Hydriding

[.

0.1 (5)

Fratting i

g (5)

'I 3

x 5

u-0.01 Unsystematic Failures (All Plants)

I (Some Hydriding And Fretting.

l Cause Frequently Unknown) g 5

PWR 0.001 j

1969 1970 1971 1972 1973 1974 1975 1976 1977 Year Figure 3-1. Fuel rod failures in LWRs for the years 1969-1977 (Garzarolli and I

Stehle,1979) 3-3 I

10 E Hydriding'PCI A PCI(Ramp incidents)

O Fretting I

~

V Corrosion By Crud S Cause Unknown w-g i

a*

T'>j 0.1 g*"..,0,'*6%

Slopes Are Due To g.

j Changing Ratios Of Old And improved

\\

'. Fuelin Service

\\

.~~

I 0.01

'O d

..f

/

~

j

. *,. 9*.

i p....

BWR 9,"i" V "f" V i

i i

i g

I 1969 1970 1971 1972 1973 1974 1975 1976 1977 Year I

10 l

O Ciad Collapsing N Hydriding A PCI(Ramp incidents) j

-G

~

O Fretting S

G Cause Unknown

_8 u.

_ _ - a, at

's, I

}

M i

j 0.1

\\

A,

,s 8

[ 0.01

,g,

'" 9 'n g

e....,

'...... g.

~.$.. +

O'

O.,'**

PWR

~~ o_,,,,,9.s'.1 O.001 1969 1970 1971 1972 1973 1974 1975 1976 1977 Year Figure 3-2. Fuel rod failure trends in LWRs for the years 1969-1977 (Garzarolli and Stehle,1979) 3-4 i

r 1

3.3.1 Fuel-Pellet Cracking I

During operation in a reactor, the steep thermal gradient in the fuel and the fracture strength of UO leads to cracking of the pellets. The resulting cracks provide pathways for radial migration of 2

volatile fission products to the pellet-cladding gap, and also increase the surface area of the fuel that could be exposed to vapor / gas and groundwater in a repository. No systematic study of the surface area I

increase of the fuel as a result of pellet cracking is available IIvwever, post-irradiation examination (PIE) on a limited number of fuel pellets indicates an increase in the crack density in the high-power fuels. Panicle sizes reported for irradiated PWR fuel with an average burnup of 28 mwd kgU range from 2 to 6.73 mm for 86 weight percent of the fuel, with less than 0.08 weight percent smaller than 0.1 mm in diameter (Katayama et al.,1980). In the same study, it has been reponed that the surface area 2

2 of the fuel increased from an estimated 38 mm mm-1 of fuel length to more than 100 mm mm-1 of fuel length as a result ofirradiation. Since the surface area and the size distribution of the fuel particles are likely to be factors that would control the release rate of radionuclides from the fuel upon contact with water, such data may need to be considered in developing source-term models.

I 3.3.2 Fuel Restructuring I

Fuel restructuring involves physical and chemical changes, including grain growth, central void formation, fission gas bubble formation and release, formation of metallic ingots, and migration of volatile compounds within and outside the fuel pellets, depending on their vapor pressures. As stated i

I earlier, LWR fuel rods typically operate at 15 to 25 kW m-1 At these linear powers, the fuel centerline temperature is less than 1200*C for a typical fuel (Guenther and Barner,1981). A review ofliterature on fuel rod performance indicates that equiaxed grain growth (restructuring) requires temperatures in the i

,g range of 1300*C to 1650*C, depending upon the fuel type and fuel rod design (Olander,1976).

5 Therefore, at typical operating linear power, little if any restructuring of the fuel pellets would be j

expected. This has been confirmed by a reported examination of PWR spent fuel with an average burnup j

of 28 mwd kgU- (Katayama et al.,1980). However, a small portion of a typical reactor core is lik.ely j

to be operated at a high enough fuel centerline temperature to result in the rectructuring of a small 1

fraction of the core load. The current trends point toward higher burnup and a higher operating temperature. These trends may require that consequences of restructured spent fuels be considered in lI developing source-term models.

3.3.3 Microstructure and Fission-Product Mobility At low temperatures (i.e., below centerline temperatures of 1000*C), fission products are expected to remain in the vicinity of the fission recoil position, since their diffusion coefficients in UO2 are typically < 10-21 ms (Johnson and Shoesmith,1988).

Therefore, limited, if any, 2

-3 microstructural changes are likely to occur in the majority of the fuel. Unfortunately, little data from direct observations are available regarding the size, composition, and location of fission-product phases in UO fuel irradiated at fuel centerline temperatures less than 1200*C. This is due, perhaps, to the difficulty in experimentally detecting submicroscopic phase segregations. Since a substantial portion of the reactor core operates in the 1000*C to 1200*C fuel centerline temperature range, additional investigations of LWR fuels operating in this temperature range may be desirable.

When fuel centerline temperatures exceed the 1200*C to 1300*C range, fission-product phase segregation begins to become significant. At these temperatures, the diffusion coefficients increase sufficiently to permit accumulation of volatile fission products in inter-and intra-granular pores, leading I

i

r

.=

te fission product accumulations on the grain boundaries and an increase in the fuel grain size. The intersection of cracks in the fuel pellets with intergranular porosity reportedly results in the release of volatile species, such as Xe, Kr, Cs, and I, to the void spaces in the fuel rod. At even higher fuel centerline temperatures of 1800*C and above, noble metal inclusions up to 10 m in diameter have been observed, and, in one study, up to 50 percent of Mo in the fuel was found in metallic inclusions (Kleykamp,1979). The remainder was believed to be distributed submicroscopically throughout the fuel.

I Yet another study reported the detection of small ingots (about 1 pm or smaller in diameter) of noble metal at the center of a PWR spent fuel (Barner,1984). In general, metallic fission products, such as Mo and Ru, and, to a lesser extent, Tc, Pd, and Rh, develop in fuel irradiated to high burnup (Yang and Olander,1981; Adachi et al.,1988). Although such high fuel operating temperatures are outside the normal operating range for the majority of the fuel in current design LWR cores, a small part of the discharged fuel could show substantial segregation of noble metal phases. A significant presence of noble metal phnes and their distribution and morphology could alter the radionuclide release rates from the spent fuel as such phases have very low solubility in aqueous solutions (Pearcy and Manaktala,1992).

g 3.3.4 Inventory and Distribution of Fission Products Of the 103 known elements, 81 are found in spent LWR fuels; of these,57 have radioactive isotopes in spent fuel. Only limited data are reported in the literature on actual distribution of the fission and reaction products within the fuel grains, at the grain boundaries, or in the fuel-cladding gap. In one of the few experimental studies on this subject reported in the literature, data and estimates are provided for selected radionuclides as shown in Table 3-1.

The results from another study, providing the calculated total fission product and actinide inventories for LWR fuel with a typical burnup of 33 mwd kgUM, are shown in Table 3-2. The data I

reported are for radionuclide inventory, taking into consideration decay times ranging from 0 (time of discharge from the reactor) to 10,000 years. It should be noted that a wide variation could be expected in the isotopic composition of the various radionuclides, depending upon the initial composition and I

enrichment, burnup, reactor operating conditions, type of reactor, and other fuel and service conditions.

The calculated results of sensitivity analyses on 10-year old fuels are shown in Table 3-3. The values are given for a number of typical fuel enrichments for both BWR and PWR fuels (Gray et al.,1992).

In an experimental investigation of the reaction of fission products at the fuel cladding inner surface, it has been reponed that Ce (and to a lesser extent, Rb) can move by vapor transport to cooler I

parts of the fuel rod, such as the cladding inner surface or the plenum region. Other fission products that can vaporize and are believed to migrate to the cladding are reported to be: Cs (Rb) in compounds; I (Br) and Te (Se) in compounds with Cs; elemental Ce, Sb, Ag, Sn, and Pb; Ba, as BaO; and possibly moo2 (Cubicciotti et al.,1976). Limited information is currently available on the quantities of these elements and compounds in the fuel and/or at the cladding surface, making it difficult to incorporate their releases into the source-term computational models.

I 3.3.5 Fuel Pellet Rim Effect The structure and fission product inventory within a few pm of the fuel pellet edge (the ' rim')

are affected by neutron flux that peaks at the pellet edge in LWR fuels. This results in higher concentrations of fission products near the fuel edge than at the fuel center (Guenther et al.,1990).

Porosity increases in the fuel at the pellet edge as burnup increases. This phenomenon has been attributed g

u I

l

E..'

I Table 3-1.

Fractional distribution of radionuclides in LWR spent fuel (Wuertz and Ellinger,1985)

Fractional Distributions Nuclides of Nuclides Similar Chemical Fuel-cladding Grain Analyzed Characteristics Gap Boundaries Fuel Matrix 137Cs Rb 0.05 0

0.95 87 135Cs 0.05 0

0.95 155Eu 147Sm 0.01 0

0.99 151Sm 0.01 0

0.99

!"Ru 93Zr 0.03 0.2 0.77 We 0.015 0.485 0.5 107Pd 0.015 0.485 0.5 94Nb 0.9 0

0.1 129; 79Se 0.015 0

0.985 25Sb 126Sn 0.015 0.485 0.5 U

U-Isotopus 0.008 0

0.992 Pu Pu-Isotopes 0.008 0

0.992 241Am Am-Isotopes 0.008 0

0.992 237Np Np-Isotopes 0.008 0

0.992 I

to enhanced fission gas release that bas been observed in high burnup fuels, normally > 40 mwd kgU-l (Einziger and Strain,1986). A comparison of the porosity (' rim' effect) for fuels of relatively low burnup,33 mwd kgU-1, and relatively high burnup,44 mwd kgU-1, is shown in Figure 3-3. It is also reported that Cs concentration shows a maximum near the fuel edge in proportion to the burnup in both fuels, although the data has not been reported (Guenther et al.,1990). The a-activity distribution is plotted on an arbitrary scale, normalized to the activity at the pellet center and refers only to the distribution 10 years after discharge from the reactor and approximately halfway through fragment corrosion tests. The phenomenon of fuel structure at the rim with a resultant enhanced release of fission products from high burnup fuel may be important in evaluating results of spent fuel leaching tests. Since the fuel pellets are small in diameter, with relatively large surface-to-volume ratio, the enhanced release of fission products from the rim of high burnup fuels may need to be explicitly incorporated in the source-term models.

I 3-7

M M

M M^

~

Table 3-2. Activity of selected radionuclides in a PWR fuel assembly irradiated to an average burnup of 33 mwd kgU-I*

(adapted from Woodicy,1983)

Activity (curies)

Radionuclide Discharge 1 yr 10 yr 100 yr 300 yr 1000 yr 10,000 yr 24IAm 5.015E O1 1.397E 02 7.740E 02 1.731E 03 1.269E 03 4.139E 02 4.734E-03 243Am 7.621E 00 7.631E 00 7.625E 00 7.563E 00 7.427E 00 6.971E 00 3.084E 00 34C 6.853E-01 6.852E-01 6.844E-01 6.770E-01 6.608E-01 6.072E-01 2.044E-01 135Cs 1.711 E-01 1.714E-01 1.714E-01 1.714E-01 1.714E-01 1.714E-01 1.709E-01 137Cs 4.786E 04 4.677E 04 3.801E 04 4.785E 03 4.783E 01 4.777E-06 0.0 237Np l.403E-01 ' l.436E-01 1.450E-01 1.914E-01 2.883E-01 4.613E-01 5.435E-01 g

23sPu 9.832E 02 1.054E 03 1.001E 03 4.970E 02 1.052E 02 4.867E-01 6.144E-20 239Pu 1.400E 02 1.424E 02 1.424E 02 1.421E 02 1.413E 02 1.387E 02 1.084E 02 240Pu 2.358E 02 2.358E 02 2.361E 02 2.352E 02 2.305E 02 2.145E 02 8.525E 01 242Pu 8.294E-01 8.295E-01 8.295E-01 8.294E-01 8.291E-01 8.281E-01 8.147E-01 226Ra 5.867E-09 1.104E-08 1.457E-07 1.145E-05. 1.142E-04 1.336E-03 5.733E-02 90Sr 3.493E 04 3.408E 04 2.729E 04 2.964E 03 2.138E O1 6.780E-07 0.0 99Tc 6.095E 00 6.124E 00 6.124E 00 6.122E 00 6.ll8E 00 6.104E 00 5.927E 00 126Sn 3.577E-01 3.577E-01 3.577E-01 3.575E-01 3.570E-01 3.553E-01 3.338E-01 235

The fuel assembly initially contained 461 kgU, enriched to 3.2% in U.

M M

M~

Table 3-3. Results of sensitivity test cases for 10-year-old fuel (Welch et al.,1990)

Ratio of Test Case to Reference Case Reference PWR llWR Standard IAw Iligh Extended.

Constant Iew Composition, Burnup Cross Enrichment Enrichment Constant burnup Cross Actinide Cross Power Nuclide grams Sections PWR PWR Flux PWR Section PWR Section PWR Level Activation Products 3dC 3.E-01 1.08 1.25 0.81 1.00 0.91 0.97 1.00

Co 2.E + 00 0.34 1.22 0.82 0.99 0.98 0.97 0.97 39Ni 4.E + 01 0.25 1.21 0.83 1.00 0.91 0.97 1.00 125Sb 1.E-01 2.35 1.23 0.81 0.98 1.12 0.96 0.94 Fission Products s5Kr 1.E + 01 0.94 0.89 1.08 0.99 0.98 0.99 0.98

Sr 3.E + 02 0.97 0.86 1.10 1.00 0.97 0.99 0.99 90Y 9.E-02 0.97 0.86 1.10 1.00 0.97 0.99 0.99 w

!"Ru 1.E-01 0.75 1.28 0.78 0.94 1.04 1.03 0.90 IU3Rh 4.E + 02 0.%

l.04 0.98 1.00 1.02 1.0'2 0.99 125Sb 1.E + 00 0.84 1.17 0.86 0.98 1.07 0.99 0.95 138Xe 4.B + 02 0.96 0.97 1.03 1.00 0.99 1.02 1.00 133Cs 1.E + 03 0.99 0.97 1.03 1.00 0.99 1.01 1.00 83dCs 3.E + 00 0.94 1.15 0.85 0.97 1.06 0.%

0.96 337Cs 8.E + 02 0.98 1.00 1.00 1.00 1.00 1.00 0.99 343Nd 7.E + 02 1.00 0.90 1.09 1.00 1.03 1.01 1.00

'#9Sm 3.E + 00 0.87 0.97 1.05 0.91 1.14 1.02 0.88 15tSm 1.E+ 01 1.06 0.99 1.03 1.00 1.21 1.05 0.98 354Eu 1.E + 01 1.03 1.25 0.79 0.99 1.06 0.97 0.99 155Eu 2.E + 00 1.05 1.25 0.80 0.99 1.14 0.99 0.99 155Gd 8.E + 00 1.13 1.25 0.80 0.99 1.14 0.99 0.99

M M

M*

Table 3-3. Results of sensitivity test cases for 10-year-old fuel (Welch et al.,1990) (cont'd)

Ratio of Test Case to Reference Case Reference MVR BWR Standard Low liigh Extended-Constant imw Composition, Burnup Cross Enrichment Enrichment Constant burnup Cross Actinide Cross Power Nuclide grams Sections PWR PWR Flux PWR Section PWR Sntion PWR Level Actinides 23 &fh 6.E-03 1.10 0.89 1.09 1.00 0.98 1.02 1.04 234U l.E + 02 1.00 0.90 1.08 1.00 0.99 1.01 1.00 235U 7.E + 03 1.03 0.48 1.72 1.00 1.13 1.04 1.00 236U 3.E + 03 1.01 0.71 1.22 1.00 0.98 0.99 1.00 23sU 9.E+05 1.00 1.00 0.99 1.00 1.00 1.00 1.00 237Np 4.E + 02 1.04 0.92 0.97 1.00 1.10 0.97 1.00 23sPu 1.E + 02 1.24 1.21 0.76 1.01 1.17 0.99 1.03 gj; 239Pu 5.E + 03 0.93 1.03 0.96 1.00 1.15 0.86 1.00 240Pu 2.E + 03 0.96 1.10 0.88 1.00 1.02 0.66 1.00 24tPu 7.E + 02 1.05 1.19 0.79 1.00 1.19 1.24 0.99 242Pu 4.E + 02 1.07 1.56 0.61 0.99 1.04 1.25 1.00 24 Am 4.E + 02 1.09 1.19 0.79 1.00 1.19 1.23 1.01 243Am 7.E + 01 1.21 1.96 0.49 0.99 1.14 1.22 1.00 242Cm 3.E-03 1.81 1.22 0.77 1.10 1.31 1.15 1.19 244Cm 1.E + 01 1.35 2.58 0.38 0.99 1.24 1.12 1.00 246Cm 7.E-02 1.48 3.94 0.25 0.99 1.28 0.99 1.00

I..-

i

,__ % ^. m m m., -N ~..w m I

~7_-,03 L. :a e s. g.; - :....

ATM 1

~

^ ~

c:

- 33.s.,,.

m-

.,. _ t

.J:1.p ^

a.:.

~w-

s, qs ;

. ~... *.

... s.

7.

-<l 13 gq.& ;g%g*g s.+; l 7. p.j_/ -

l

<gg7wp. 7-%.N y

-l

-4 1. t,.

4 i O A 7;.7P L U /.jW

~

, 7 3;

- r.

e 3

g.,.

,'.e..

~40p. n W.= # W Y..-hs,I' a y'

? ~.-

u-u.=.x.. ; ew -

?.e..w:

_;.x. ;

.z

..ATM 10. 4 ? '.. * /m..

-n

. '. '. : e.s :

l

.% ~

...,n.

a s

_.. - e J. " ;, : y-y

... =.

r

. L 6 l ' * :. '.J.*

.. o_

.-,...-..,-7.,..

.- f~ " ' ;t *.*.*. *. 9,,.. ". J.a *

-~

I

..-s.-

.e.cp."

a.s.,:

, y.

.,.....a

-4. w.._s.,,.

.. e.e j i:

,p.sp,. g, u.

+ - r

~

,.,.y-

.~

-M,... G' ;.

I

., d Mge:

,' j f.

J-40 rn g,3.. '

y.

x.

Figure 3-3.

Typical porosit on ATM-103-33 mwd kgU-1;y near the fuel pellet edge (Burnup:on ATM-104-44 et al.,1990) 3.3.6 Fission Gas R.elease and Pressure Typical end-of-life (EOL) room temperature pressures for BWR and PWR fuel rods have been reported to be approximately 1.5 MPa and in the 3 to 4 MPa range, respectively (Johnson et al.,1982).

I Other factors, such as storage temperature and actinide decay, are also likely to affect the gas pressure in an intact fuel rod. Calculated pressure increases, as shown in Figure 3-4, have been reported due to He production from actinide decay over a 10,000-year period (VanLuik et al.,1987; Johnson and Gilbert, I

1983). These increases have been calculated for temperatures of 380*C and 100*C, which reportedly represent a conservative peak design temperature (380*C) for the early part of spent fuel life in a geological repository and a typical temperature (100*C) calculated for the end of the contamment period.

I The calculations are based on a PWR fuel rod with 36 mwd kgU-l burnup and an assumed 100 percent He release.

The isotopic composition of the gas in the plenum, gap, and cracks between the fuel fragments is difficult to estimate accurately because of power history dependence and radioactive decay. However, prepressurization of the fuel rods with He typically leads *.o a reduction in the fission gas release by as much as an order of magnitude, as shown in Figure 3-5 for pressurized and unpressurized BWR and PWR fuel rods. Most of the gas is composed of Xe, Kr, Ar, and He, with small contributions from other gases such as H, CO, N, 0, and volatile organic compounds. In an investigation reported in the 2

2 2

2 literature, He cor.stituted :foout 98 percent, and Xe, Kr, and At together accounted for abcut 1 percent, 3-11 I

g 1.0 E41 380*C 100* C E

E 1.0Edo

[

R 3ti5 e

5:

y 1.0 E-01 1.0 E-02 3

104 0

101 102 10 10 Tkne,y Figure 3-4. Calculated pressure increase caused by helium production from actinide decay (Johnson and Gilbert,1983)

{

while all the remaining gases constituted about 1 percent of the total volume (Barner,1984). Apart from the isotopic changes that result from radioactive decay, the amount of volatile fission products in the free void volume of an intact fuel rod is expected to remain constant after discharge, as long as the fuel

{

temperature remains below 1000*C (Fish and Einziger,1981). It is further reported that nearly all of the fission gas is retained in the fuel pellets, even after 10 years of spent fuel pool storage, and this would probably be valid for dry storage as well (Fish and Einziger,1981). Additional fission gas would not be expected to be released under repository storage temperatures unless the fuel cladding ruptures and UO oxidizes to U Og or higher oxidation states that rearrange and open the lattice structure, creating 2

3 a much larger fuel surface area from which the gas could be released. Oxidation studies of spent UO 2 fuel have shown that cumulative release of 85Kr and the change in the oxygen-to-metal (uranium) (0/M) ratio track each other in Advanced Gas Reactor (AGR) fuel oxidized above 250 C (Bennett et al.,1987; Williamson and Beetham,1987). In a study on oxidation of spent AGR fuels, it is reported that even after complete oxidation of UO to U 0s, a large fraction of gas remains in the fuel matrix. The amount 2

3 85 reported for release of Kr is between 7 and 30 percent (Wood et al.,1985). No measurements of release have been made at temperatures likely to dominate the majority of the repository performance calculational thnes. Although data from comparable tests on LWR fuels (BWR and PWR) as they oxidize

[-

from UO,, to U Og, which could be used for modeling gaseous release due to fuel oxidation, are not 3

available, it is believed that similar trends will hold (Einziger,1991). Such a scenario for the oxidation of spent LWR fuel is a distinct possibility in an unsaturated repository during the period when the spent

[

fuel temperature is still high (e.g., in the range above 250*C).

3-12

25 I

O sw Rock Pom BwR gD d Connec5cutYankee QgD PWR 20 Unpressurized Rods c

0 or n BwR I

D O Mahe Yankee PWR Y

h Oyster Creek BWR g

I 7 Peach Bottom 2 BWR g

e 15 Y

I o

3 5

e o

I j

10 V

I O

O 5

A 00 o

O

'MN O CI t

OC I

0 10 20 30 40 50 Burnup, mwd

kgU-T I

2.5 O ceivert cune 1 PWR O Ooonee 2 PWR Pressurirod Rods 2.0 I

/

4 i

i

.E j

1.5 o

i

~3 j

.a:

j O

f j

1.0 D

O h

O o

8 0.5 aO o

@R 0

8 a

j 6

I I

O 0

10 20 30 40 50 l

l Burnup, mwd

kgu-1 Figure 3-5.

Comparison of fission gas release from unpressurized and I

pressurized LWR fuel rods as a function of the fuel burnup (Ocken,1982) 3-13

'I

E..

4 SPENT FUEL AND CLADDING CHARACTERISTICS, STORAGE, AND INVENTORY I

4.1 RADIOACTIVfIT ASSOCIATED WITH CLADDING The radioactivity in spent fuel Zircaloy cladding results mainly from (i) the formation of 125 activation products such as Sb and "Co by neutron irradiation of alloy constituent (Sn) or impurities, 137 (ii) the recoil-implantation of fission products such as Cs and IM 134 Ru, and radionuclides, such as Cs I

and IMEu, produced by neutron capture of recoil-implanted fission products, (iii) the capture of H 3

produced in fuel and/or reactor coolant, and (iv) the occlusion of actinides and fission products into the inner surface layer of cladding during irradiation. In addition, activation products ("Co,54Mn) from I

piping and structural materials in-core get incorporated in the crud deposited on the fuel rods. The results of a study on Zircaloy-4 cladding of spent PWR fuel with burnup in the 7 to 40 mwd kgU-l range are summarized in Table 4-1.

4.1.1 Radioactivity as a Function of Fuel Burnup I

Figure 4-1 shows the activities of a-emitters on the inner and outer surfaces of cladding as a function of burnup of fuel. The activity of a-emitters on the inner surface of cladding increases with the 3.1 power of fuel burnup, while that on the outer surface increases with the 1.3 power. On the cdier I

hand, the gross a-activity of the fuel, as shown in Figure 4-2, increased with the power of 2 of the burnup of fuel. These results indicate a preferential accumulation of a-emitters on the inner surface relative to the outer surface. Figure 4-3 shows the activities of representative S-and g emitters in spent nuclear fuel claddinos as a function of burnup of fuel. The fission products 137Cs, I

,3H, and the I

E activation product Co in the claddingincrease proportionately to the first power with burnup of fuel, ll4 whereas the radionuclides Cs and Eu increase with the power of 2 of the burnup of fuel. These g

results suggest that, although single events such as fission reaction of 235U produced directly the former g

radionuclides, the successive reaction with neutrons, that is, the neutron capture after fission reaction, pMarticipated in the formation process of the latter radionuclides. As a result, the formation of 13#Cs and Eu appears to show seconderder dependence on the burnup.

5 4.1.2-Distribution of Radionuclides The activities of radionuclides in the etching solutions obtained by step-wise etching of the cladding from the inner surface are shown in Table 4-2 and Table 4-3. The data represent cladding from PWR fuel with 29.4 mwd kgU-l burnup. Table 4-2 shows that more than 98 percent of the fission 137 1

products, such as Cs and *Ru, are distributed within about 10 m from the inner surface, while the activation products ("Co,125Sb) are distributed homogeneously in the interior of the cladding except in the oxide film on the outer surface, where they are relatively enriched. The heterogeneous distribution 3

of H was found in the cladding at high burnup, although it was virtually homogeneous in the cladding 3

with low burnup. This implies that segregation of H occurs as a hydride in the cladding at high burnup.

j Since Zircaloy cladding amounts to almost 300 g per kg of UQ fuel, and it has a considerable surface j

area, its leaching properties for the release of radionuclides may be of interest in developing source-term models. Table 4-3 shows that a-emitters are principally present on the inner and outer surfaces, and only a small amount is distributed in the interior of the cladding. The a-emitters concentration ranged from j

1.3 to 36 kBq g-1, except for the inner (< 3.7 ym in depth) and the outer (> 617 pm in depth) surface K

a 4-1

g g

g-- g gg g.

Table 4-1. Concentration of radionuclides in the spent fuel claddings (adapted from Ilirabayashi et al.,1990)

Concentration of Radienuclides (Bq

g-1 cladding)

Burnup 11 Mn

'"Co I"Ru 125 84 83d 337 (mwd

kgU-8) 3 Sb Cs Cs 8"Ce 85dEu 6

6 5

6 7

5 6

5 4

6.9 3.41 x 10 5.92 x 10 3.52 x 10 1.74 x 10 1.04 x 10 3.04 x 10 3.06 x 10 9.79 x 10 3.26 x 10 4

5 6

7 5

6 5

8.3 6.64 x 10 4.15 x 10 2.11 x 10 1.17 x 10 4.40 x 10 4.58 x 10 9.98 x 10 4.58 x 104 7

5 5

6 7

6 6

5 14.6 1.12 x 10 1.11 x 10 7.48 x 10 6.60 x 10 2.14 x 10 1.74 x 10 9.74 x 10 1.34 x 106 2.22 x 10 7

5 5

s 7

6 6

s 5

15.3 1.67 x 10 1.16 x 10 7.62 x 10 5.21 x lo 2.23 x 10 1,73 x10 9.85 x 10 1.85 xlo 2.12 x 10 7

5 5

6 7

7 6

5 21.2 2.24 x10 1.14 x 10 9.10 x 10 6.19 x10 2.43 x 10 3.05 x los 1.37 x 10 2.19 x 10 4.01 x 10 7

5 6

6 7

5 29.4 3.19 x10 1.41x10 1.30 x 10 9.51 x 10 3.44 x 10 6.53 x 10 2.32 x 10 8.62 x 10 7

5 6

7 7

6 7

s 6

32.1 3.40 x 10 1,48 x 10 1.45 x 10 1.10 x10 3.73 x10 7.96 x 10 2.59 x 10 2.33 x lo 1.04 x 10 b

7 5

6 7

7 6

7 3.48 x 10*

1.03 x 106 33.6 3.59 x10 1.33 x10 1,42 x 10 1.01 x 10 3.51 x 10 8.13 x 10 2.71 x 10 7

5 6

7 7

6 7

6 34.0 3.81 x 10 1.53 x 10 1.51 x 10 1.17 x 10 2.78 x 10 9.19 x 10 2.95 x 10 1.21 x 10 7

5 5

7 7

7 6

38.1 3.77 x 10 8.86 x 10 1.46x10 4.37 x 10 1.03 x 10 3.li x 10 1.36x 10 7

4 5

7 7

6 38.7 3.50 x10 8.69 x 10 8.08 x10 1.43 x10 4.30 x 10 1.06 x 10 3.08 x 10 1.42 x 10

.~

104

.Y j

- Inner surface N

~

E E

3 g 10 e

I E

5 O

=

?

l 4

o i

~

O 5

E O

102 O

I

..O,,

e 104 105 Bumup of fuel, mwd + kgU-1 Figure 4-1. Gross activities of a emitters on the cladding surfaces as a function of fuel burnup OIirabayashi et al.,1991) 109 Pu I

Gross a::twity 108 7 l5 P

p E

~

/

'l g

5 Cm e

{10 7

x 2

~

x 2

/

ss s=

lI

~~

f 104 "I

1 10 100 i

Bumup of fuel, mwd. kgU-1 i,

Figure 4-2. Elemental contribution to a activity in the spent nuclear fuel as a function of fuel burnup QIirabayashi, et al.,1991) 4-3

106 12sSb 105 i"

}

~

3 o

j H

g 134cs k3

- 137Cs

~

d 106Ru f

/

k

?

W

/

og soca 2

103 I

~

f f !? f ff f

1 f !I ff' 1

10 100 Bumup of Fuel. mwd kgirl Figure 4-3. Activity of # or y emitters in the spent fuel claddings GIirabayashi, 1991) layers of the cladding. The average concentration was found to be about 2 kBq g-1 (about 50 nCi g-1) in the interior of cladding. Very limited leaching data on spent fuel cladding are available. In one of the studies, the leaching behavior of Zircaloy cladding from spent PWR fuel was investigated (Brodda and Merz,1984). Two types of Zircaloy cladding with different pretreatments were available for j

experimental studies. The major characteristics of the two types of cladding specimens are given in Table 4-4. The specimens were boiled in a reflux condenser for 3 hours3.472222e-5 days 8.333333e-4 hours 4.960317e-6 weeks 1.1415e-6 months using 50 mL of the solutions. The solutions were filtered afte-r cooling and their radioactive inventories measured. Double distilled water 01 0), 7 M HNO,1 M NaOH, Portland cement (PC) solution, alumina cement (AC) solution, and Sorel 2

3 cement (SC) solution were used as leachants. 'Ihe leached fraction was determined for six y-emitting isotopes and two actinides. Table 4-5 show: the results, in terms of the percent leached, for the L

7-emitting nuclides. It is evident that the soluble species, Cs, dissolves relatively well in both alkaline and acidic solvents, in contrast to other elements studied. Table 4-6 gives the initial levels of the e

actinides 23*Pu,240Pu,and Cm and their behavior under successive leaching with HNO, NaOH, and 24d 3

KS0.

Figure 4-4 shows the radionuclide distributions within the Zircaloy cladding for the 2

7 experimental series. The results are divided into 3 groups: (i) the actinides that are concentrated in a thin surface layer, (ii) the diffusing species, whose highest concentration is in the inner surface layer but which also penetrate deeper into the material, and (iii) the activation products that are, panly, located on the external surface as corrosion products, but which otherwise display a homogeneous distribution.

4-4

{

v mw Tabic 4-2. Distribution of the main # and y emitters in spent fuel cladding with 29.4 mwd kgU-l burnup and 5 years cooling (Ilirabayashi et al.,1991)

Rad *w> activity (Ilq

g-I cladding)

Depth 7

(pm) 11 "Co

%r I"Ru 125Sb 83dCs

137c, 144c, 147Pm 3"Eu 7

8 8

8 2.93 x 10' 3.31x10 O - 3.7 9.83 x 10 1.14 x104 2.2x10' 8.9 x10 1.02 x 10 8.3 x10 8

2.2 x10' 1.20 x 108 7

5 1.1 x 10' 7.2 x 10' 8.25 x 10 4.07 x 10 7

8 1.33 x 10' 1.24 x 10 8.5 x10 4.1 x10 8

8 7

3.7 - 5.4 5.01 x 10 9.0 x10 7

6 6.2 x 10' 3.30 x 10 5.59 x 10 9.5 x 10 4.43 x 10' 2.64 x 10 8

7 7

7 16 x 10' 6.7 x 106 5.4 - 6.8 3.07x 10 1.8 x10 7

5 7

7 s

7 6.8 - 10 1.54 x10 9.7 x10 5.0 x10 2.13 x 10 7.1 xlo 2.22 x 10 9.0 x105 7

7 6

7 5

10 - 20 4.80 x 10 1.29 x 106 4.9 x106 3.41 x 10 4.80 x 10 1.75 x 10 7.0 x 10 i

7 6

6 7

5 6

5 20 - 80 6.96 x 10 1.19 x 10 1.11 x 10 3.21 x10 8.8 x10 3.39 x 10 1.46 x 10 l

l 7

6 5

7 5

p 80 - 135 6.39 x 10 1.11 x 10 6.5 x10 3.15 x 10 1.54 x 10 5.6 x105 1

ta 135 - 189 4.55 x 10 1.13 x 10 1.03 xlo 3.24 x 10 5.3 x 10 1.90x 10 l

7 6

s 7

4 5

7 189 - 241 5.51x10-1.15 x106 7.0 x 107 5

7 4

6 3,4 x go?

7.7 x10 l

4.0 x10 3.42x 10 4.7 x10 1.59 x105 5

2.8 x10 4

6 4

5 7

5 4

5 5

241 - 385 3.67 x10 1.22 x 10 6.9 x10 2.9 x 10 3.42 x 10 2.5 x 10' l.11 x 10 2.8 x10 1.3 x10 7.1 x10 7

6 5

7 4

385 - 524 1.61 x 10 1.23 x10 2.9 x 10 3.42 x10 3.4 x10

)

8 6

5 7

8 524 - 617 1.03 x 10 1.21x10 1.7 x 10 3.28 x 10 3.0 x10 7

7 5

6 6

7 6

617 - 620*

7.41 x 10 1.59 x 10 8.9 x10 6.3 x10 1.45x105 9.0 x10 1.58 x 10 4.2 x10 3.9 x 10 2.5 x10

Outer surface layer

m v

m.

rm w

m m

W Tabic 4-3. Distribution of the main rr emitters in spent fuel cladding with 29.4 mwd kgU-I burnup and 5 years cooling (Ilirabayashi et al.,1991) l Radioactivity (Bq g-I cladding)

Depth (pm)

Grms a U

Np Ib Am Cm 6

0 - 3.7 1.97 x 10 1.6 x 101 6 x 100 5

5 9.7 x10 3.93 x 10 5.12 x 105 4

0 0

3.7 - 5.4 3.6 x10 4 x10

< 4 x 10 1.48 x 104 7.43 x 103 3

8.7 x 10 5.4 - 6.8 3.03 x 10'

< l.6 x 10

< 5 x 100 0

4 1.38 x 10 4.7 x103 7.6 x 103 6.8 - 10 7.6 x103 10 - 20 8.7 x 103 20 - 80 3.5 x 103 k

80 - 135 2.06 x 103 135 - t89 1.52x 103 189 - 241 1.47 x 103 3

241 - 385 1.39 x 10

<5 x10 2

<5 x100 6.1 x 101 2.69 x 102 3.15 x 102 3

385 - 524 1.49 x 10 l

3 524 - 617 1.61x10 617 - 620*

2.20 x 105 5 x 101

<2 x 100 1.45 x 105 250x104 4

3.33 x 10

Outer surface layer I,

g-

__i

t3 5

Table 44. Characteristics of cladding specimens investigated (adapted from Brodda and Merz,1984)

Specimen Identification Type K W U (P-1)

WAK (P-2a and P-2b) l Initial enrichment 3.1 %

3.02 %

Burnup 30 mwd kgU-l 28.2 mwd kgU-1 Mean power density of fuel rod 180 W cm-1 200 W cm-1 j

Discharge from reactor 1977 1975 Reprocessing date Not applicable 1977 l

Cladding hull material Zircaloy-4 Zircaloy4 t

Separation fuel-cladding hulls Mechanical Chemical i

l I

From that investigation, it can be concluded that there are two main sources of radioactivity.

He first is surface contamination or surface nuclide implantation due to nuclear recoil effects-the i

observed leaching of various nuclides by different aqueous solvents indicates that the radioactivity must g

be, in part, pure surface contamination. However, some fraction is also firmly implanted in the oxide 5

layer. Although the actinides. like the rare earths, display very low diffusion rates in Zircaloy, the l

opposite is true of some fission products. He depth distribution measured for Ru and Cs leads one to suppose that similar diffusion behavior may also be valid for other chemically related elements. It thus I

becomes clear that a surface decontamination of the Zircaloy cladding would almost completely eliminate the actinides but not all fission nuclides. The second source is activation products of the cladding which are distributed fairly evenly through the thickness of the material-the radionuclides "Co and 1 Sb are I

characteristic of this group. Based on the results of the investigation, one could conclude that cement l

could improve the radionuclide leaching resistance-that is, that actinides cannot be put into solution by alkaline solvents and thus cannot be made mobile. Therefore, embedding the Zircaloy cladding in cement c

for disposal may be beneficial. Cement, however, does not contribute to permanently improving the F

Icaching resistance of alkaline and alkaline earth elements nor, presumably, that of Ru. It may just delay their release.

I In developing packaging concepts for Zircaloy clad spent fuels, the level of H and 85 3

Kr may L

also play a role. Herefore, the generation of hydrogen gas by radiolysis, which can lead to an increase in tritium release, must also be taken into consideration in estimates of the distribution of the canister 3

failures as a function of time. The H is rather firmly bound into the Zircaloy matrix in the form of

~

zirconium tritide and is released only at relatively high temperatures (800 to 1000*C). On the other hand, zirconium tritide exchanges its hydrogen isotopes in contact with aqueous solutions or hydrated solids, such as cement products. An isotope exchange process of this type can lead to an increase'd H 3

L release from the cementee Zircaioy clad fuels. The behavior of the inert gas s5g,, w33;3 enters 13, Zircaloy cladding by means of recoil effects, is especially interesting. A mean 85Kr inventory of 62 Ci per gram of Zircaloy wa.3 measured for PWR cladding with a burnup of 31 mwd kgU-l (Brodda and 4-7 I

i]

Table 4-5. Percent release of y-emitting nuclides from Zircaioy cladding specimens in different teachants (Brodda and Merz, 1984)

Radionuclide Leached

(%)

Initial Ilull Radioactivity Type Nuclides 0Ci g-3) 11 0 IINO NaOli PC AC SC 2

3 j

l P-1 134Cs 181 7.8 10.9 15.4 9.8 13.8 12.7

17Cs 936 7.2 11.0 14.9 9.9 13.8 12.7 I"Rh 414 ND 21.4 1.1 ND ND 1.7 I

l 154Eu 33 ND 0.4 ND ND ND ND 125Sb 1234 0.8 2.3 3.5 0.2 0.9 0.5 "Co 69 ND 0.3 1.5 ND ND ND P-2a 334Cs 41 1.5 12.4 3.7 2.4 3.2 1.3 3

do 137Cs 498 1.5 12.5 3.7 2.3 3.0 1.4 1*Rh 13 ND 99.5 1.8 NM NM NM 154Eu 14 ND 2.8 ND ND ND ND 125Sb 222 0.9 30.2 7.7 0.1 1.0 1.8 l

l

Co 26 ND 40.7 2.9 ND ND ND l

l P-2b 134Cs 0.4 49.8 81.9 56.I NM NM NM l

337Cs 3.8 75.8 95.6 86.1 NM NM NM l

1MRh 2.5 24.0 99.5 2.7 NM ND NM l

154Eu ND ND ND ND ND ND ND 125Sb 45 23.2 28.3 10.3 ND 28.0 NM "Co 8

9.1 48.4 4.3 NM ND NM ND = Not Detectable NM = Not Measured l

PC

= Portland Cement solution AC

= Alumina Cement solution SC

= Sorel Cement solution I

l

\\

l

I Table 4-6. Percent release of actinides from Zircaloy cladding spximens in different leachants (Brodda and Merz,1984)

Initial Radioactivity Leached Amount I

4Ci g-1)

(%)

Ilull Type 239,2g 2"Cm Leachant Use 239.2ePu 2MCm I

P-1 150 260 HNO 60 46 3

K S Oy 40 54 22 NaOH ND ND KS0 57 52 2 7 P-2a 1100 570 HNO 90 40 3

KS0 10 60 22 7 I

NaOH ND ND K 5 Oy 59 100 22 P-2b 440 10 HNO 98 100 I

3 KS0 2

ht 22 7 NaOH ND h3 ND = Not Detectable 100 I

I = Actinides, Pu and Cu ll = Diffusible species, Ru and Ce 111 = Activa *aon products, Co and Sb 80

[

I E

$ 60 2

No I

E j

40 g--__x u

s s

I h

\\

'e ll

/

s s

20

\\

's

's I

i

'x

's

' x 's 1

l 100 80 60 40 20 0

Residual mass of cladding, %

I'igure 4-4. Radionuclide distributions in Zircaloy cladding specimens (Brodda and Merz,1984) 4-9 I

g..-

Merz,1984). This corresponds to approximately 0.1 percent of the total inventory in the fuel rod. It has also been shown that the 85Kr activity is almost completely restricted to the inner corrosion layer of the cladding tubes, which is approximately 10 pm thick.

4.2 RADIOACTIVITY IN SPENT FUEL When fuel is irradiated in a reactor, three general groups of radioisotopes are formed: (i) fission products, (ii) actinides, and (iii) activation products. More than 350 nuclides have been identified as fission products, many of them with very shon half-lives. The exact quantities depend on the irradiation I

history of the fuel and the time after discharge. The concentration of any isotope can be computed, in principle, using the following differential equation, Eq. (4-1) (Tsoulfanidis and Cochran,1991):

l N

N dN,Jdt = [ 1,, X, N, + & [ f;& a N i

1

/-t 1-t (4-1)

- {X, + & a, + r,) N, + F, l i = 1, M where Nj atom density of nuclide i

=

number of nuclides in the chain containing uclide i M

=

lg fraction of decays of nuclidej leading to the formation of species i

=

A, radioactive decay constant of species i

=

position and energy-averaged necen flux

=

f,3 fraction of neutron absorption by nuclide k leading to formation of species i

=

at average neutron absorption cross section of nuclide k

=

r, continuous removal rate of nuclide i from the system

=

F, continuous feed rate of nuclide i; if nuclide i is produced by fission,

=

i F, = Y,qip, where Y, = fission yield of nuclide f.

Equation (4-1) is not linear if either the flux, 4, or any of the cross sections are functions of time. The activity of the nuclides produced by the activation of Zr in the cladding and of Ni and Co isotopes found in the steel used in control rods and in fuel assembly spacer grids should be added to the activity from fission products and actinides.

I To calculate the total activity in spent fuel as a function of time, Eq. (4-1) must be solved for all isotopes of interest and the results summed. A widely used code that performs this task is ORIGEN (Croff,1983; Bell,1973; Klein,1987; Bowman,1992). Analytical fits for ORIGEN code have been

.g me E

g..-

developed to give the activity as a function cf time after discharge (Locke,1975). The calculations of the code have been validated against experimental data on spent fuel radionuclide inventories (Heeb et al.,1990).

4.3 CIIARACTERISTICS OF SPENT FUEL 238 235 Irradia'.ed fuel still contains most of its original U, about one-third of its original U,

almost all the fission products, all the transuranic (TRU) isotopes, and many activation products. Tables 4-7 and 4-8 summarize the most important isotopic features of typical LWR spent fuels. The volume of I

wastes generated during the service life of LWRs is given in Table 4-9.

4.4 STORAGE OF SPENT LWR FUEL I

Prior to permanent disposal in a repository, the spent fuel is expected to be stored for some years (perhaps 10 yeaa or more) to allow the assemblies to cool down as a result of dissipation of the decay heat. Two principle methods are available for storing the spent fuel for extended periods of time, namely, wet storage in water pt,als and dry storage in air or gases (EPRI,1986; Bailey et al.,1986; EPRI,1984; Johnsor.,1979). Water pool storage has been used in the U.S. since the first reactors were built in 1943 (Johnson,1977). However, experience with dry storage, an alternative to extended wet storage, is rather limited. The fuel may be stored unconsolidated or consolidated. By defir.ition, unconsolidated fuel means intact assemblies; consolidated means the fuel rods are removed from the fuel I

assembly and placed in a grid with cbser spacing than that of an intact assembly, or the rods are placed in a close-packed array inside a canister (Zacha,1988; Johnson,1986; EPRI,1989). Volume savings of 2:1 by consolidation have been demonstrated by many utilities (Zachy,1988). The advantages of fuel I

rod consolidation are obvious: (i) storage capacity would be almost doubled, and (ii) the number of spent-fuel shipping casks can be halved.

I In the wet storage mode, the decay heat from the spent fuel is removed by deionized water (DIW) at a temperature below 40*C. Zircaloy cladding does not experience any significant additional corrosion or hydriding under such conditions compared to that experienced while in core. In the dry storage mode, the decay heat is removed by using air or an inert gas (usually helium or nitrogen) under I

forced convection. The dry storage facility is simpler and cheaper to maintain mmpaeed to wet storage.

On the other hand, the dry environment has the potential of producing such problems as further fuel cladding oxidation, increased cladding stresses ar.d creep deformation as a result of rod internal pressure, I

and volume expansion of the fuel due to air permeating through any pinholes and incipient cracks in the cladding. These possible spent fuel and cladding alteration modes could be quite accelerated under dry storage conditions, since the temperatures are much higher than in wet storage. Temperatures in the range I

of 300 to 400'C are being considered for the extended dry storage of spent fuel for periods up to 100 years following discharge from the reactor. Because of the stated reasons, any extended cry storage may require evaluation of any additional process of materials degradation or alteration of the cladding I

properties which may influence its subsequent behavior in a repository. Ieng-term behavior of extended wet stored spent fuel may need to be modeled differently from assemblies that experience dry storage following discharge from the reactor.

I Allowable long-term storage temperatures and times for dry storage of spent LWR fuels will depend upon a number of factors including cladding stress levels, fuel type and assembly design, materials condition of the cladding at the time of discharge, decay heat history of the spent fuel, and heat 4-11 I

4 3.

I Table 4-7. Radioactivity from principal fission products in irradiated fuel * (Tsoulfanidis and Cochran,1991)

E l

Activity per year (1 x 10' Ci)

Nuclide IIalf-life (at discharge) i 3H 12.4 y 1.93 x 10-2 85Kr 10.76 y 0.308 "Sr 27.7 y 2.11 95Zr 65.5 d 37.3 6Ru 368 d 14.8 125Sb 2.71 y 0.237 12'I 1.7 x 10 y 1.01 x 10-6 7

f31 1 8.05 d 23.5

!We 5.27 d 43.9 l

13*C3 2.0?6 y 6.7 137Cs 30.0 y 2.94 i

Ce 284 d 30.2 2'4 15tSm 87 y 3.41 x 10-2 154Eu 16 y 0.191 155Eu 1.81 y 0.204

,E 1

The fuel ir assumed to be from a 1000-MW (electric) PWR, with an initial enriclunent 3.3%

in 235U and operated for 3 y with an 80% capacity factor.

dissipation capacity of the storage installation. A recent investigation of the potential failure modes of Zircaloy fuel cladding during dry storage, covering stress / creep rupture, stress corrosion cracking (SCC),

and cladding tupture due to delayed hydride cracking (DHC), has identified stess/ creep rupture as the likely dominant failure mode (Gilbert et al.,1990). The internal pressure from helium and fission gases within the fuel rods are the sources that may lead to such a failure, if pressures and temperatures are sufficiently high. However, another study concluded that spent PWR fuel can be stored safely for a period of 1,000 years +mder dry storage conditions at a temperature of 400*C (Einziger et al.,1982).

i 4.4.1 U.S. Fp t M Inventory The eldest W

>ent fuel currently in storage was discharged in 1968, whereas the oldst PWR spent fuel in stor as dischac;,ed in 1970 (U.S. DOE,1982). Spent fuel discharged in prior 4-12 I

Table 4-8. Radioactivity from principal actinides in irradiated fuel * (Tsoulfanidis and Cochran,1991) l IIalf-life Activity 1.er year Mass or Concentration Nuclide (y)

(Ci)

(kg kgU-1) l 2"U 2.47 x 10 19.4 0.12 x F 5

3 236U 2.39 x 10 7.22 4.18 x 10-3 7

237Np 2.14 x 10 14.4 0.75 x 10-3 6

236Pu 2.85 134 9.20 x 10-9 23sPu 86 1.01 x 105 0.22 x 10-3 239 3

5.28 x 10-3 Pu 24,400 8.82 x 10 29u 6,580 1.30 x 10' 2.17 x 10-3 241 6

1.02 x 10-3 Pu 13.2 2.81 x 10 5

242Pu 3.79 x 10 37.6 0.35 x 10-3 3

0.05 x 10-3 241A:n 458 4.53 x 10 243Am 7,950 477 0.09 x 10-3 242Cm 163 d 4.40 x 105 4.90 x 10-6 244Cm 17.6 7.38 x 104 3.30 x 10-5 l

Fuel burnup is 33 mwd kgU-1 in a 1000-MW (electric) PWR, and 150 days after discharge.

years has been reprocessed. The total inventory of LWR spent iael in storage in the U.S. as c f the end

' 1990, was 77,398 assemblies. These assemblies represent initial U content of approximately 6

l 2 x 10 kg, of which 38 percent are from BWRs and 62 percent are from PWRs (U.S. DOE,1992b).

Table 4-10 shows the inventory breakdown for BWR and PWR spent fuel discharged from 1968 through 1990 (U.S. DOE,1992b). Annual spent fuel discharges and fuel burnup are shown in Table 4-12. The majority of spent fuel is stored in water in storage pools at reactor sites, but there are 404 assemblies in E

dry storage (U.S. DOE,1992b). Of the 114 reactors expected to be in operation in the year 2000, as many as 26 reactors appear to require expansion above current pool maximums prior to the year 2000.

7 The current inventory of the spent fuel assemblies represents less than one-third of the spent fuel scheduled for disposal in the first geologic repository in the United States.

A computerized database, called the Characteristics Data Base (CDB), is currently available which provides additional information about the discharged fuel assemblies (U.S. DOE,1988a; 1988b; Notz et al.,1992). Another database that provides the quantities of spent fuel, radioactive waste inventories, and projections for high-level, low-level, and transuranic wastes both from the commercial 4-13 i

m m

m' W m

m m

M 'm m

M M

W-

^

Tal>le 4-9. Lifetime radioactive was% generation from a PWR and a BWR* (adapted from Tsoulfanidis and Cochran,1991) m Refercrxe PWR,1000 MW (electric)

Reference BWR,1000 MW (electric)

Volume Radioactivity Volume Radioactivity 3

3 Waste Type (m )

(un3ecayed curies)

(m )

(undecayed curies)

Once-through fuel cycle wastes Mill tailings 4.353E+06

3.710E+03 4.867E + 06 4.149E + 04 LLW from u anium conversionb 3.41 lH + 02 9.813E + 03 3.814E +02 1.097E + 04 LLW from uranium enrichment

1.328E + 02 9.716E+ 03 1.365E + 02 1.080E + 04 LLW from fuel fabrication 3.063E + 03 7.288E + 00 4.110E + 03 9.781E + 00 LLW from reactor power generation 3.032E+04 2.866E + 04 5.217E + 04 7.956E + 04 Reactor spent fuel 5.213E + 0::

3.270E + 09d 6.996E + 02 3.342E + 09d Decommissioning wastes LLW l.510E + 04 1.057E + 05 1.640B + 04 2.532E + 05 Greater than Class C wastes 1.130E + 02 4.070E + 06 4.070E 4 01 5.300E + 06 A

'IUTAL j

4.403E + 06 3.274E+09 4.941E,' 06 3.348 E + 09 Waste generated from 40 y of reactor operation and 26,000 MW (electric)

y s

Read as 4.353 x lo b

Applies to the fluorination /fractionationprocess Applies to the gaseo ts diffusion process d Based on activity levels measured 1 y after reactor discharge. For the PWR, these levels are based on a burnup of 33 mwd kgU-3. Activity levels for the BWR are based on a burnup of 27.5 mwd + kgU-l.

I

e I

Table 4-10. Spent fuel discharged during 1968-1990 from commercial LWRs in the United States (U.S. DOE,1992b)

I Nurr.ber of Assemblies I

Stored at Stored at Off-Site Reactor Type Reactor Site Storage Facility Total BWR 42,338 2,955 45,293 PWR 31,612 493 32,105 TOTAL 7? f 0 3,448 77,398 Amount of Uranium (kg x 10-3)

BWR 7,673.1 553.9 8,227.0 PWR 13,494.8 192.5 13,687.3 TOTAL 21,167.9 746.4 21,914.3 lI l

and defense activities is also available (Klein,1987). This database, called the Integrated Data Base (IDB), provides updated projections of the wastes in the U.S. through the year 2020. The information I

available in the CDB includes the details about the physical, radiological, and quantitative characteristics i

of the spent fuel organized by as;embly class, type codes, length and width of the assemblies and

'g individual fuel rods, number of fuel rods, type of material used for cladding, and other parts of the 5

assembly. The radiological characteristics have been generated using ORIGEN code (Parks,1992; Welch et al.,1990; Welch and Notz,1990). Included in the database is the information related to the burnup and number of damaged and leaking assemblies (U.S. DOE,1992a). The summary of data related to I

defective BWR and PWR spent fuel assemblies, as of the end of 1988, is shown in Tables 4-12 and 4-13, respectively (Moore et al.,1990). These data clearly show high defect rates for certain assembly designs.

As shown in Table 4-10, GE Model 2,7 x 7 fuel design assemblies operated in GE Model 2/3 BWRs show spent-fuel defects in 1469 out of 6719 assemblies, with a defective rate of 21.9 percent. The same design assemblies for GE Model 4/5/6 BWRs show 385 defective assemblies out of 1142 irradiated assemblies, with a rate as high as 33.7 percent. One conclusion that can be drawn from the data is that operating / service and the type of reactor could have some influence on the condition of the assemblies at discharge. Discharged assembly defect rates for PWR assemblies, as shown in Table 4-11, are much lower. Although, modern day assemblies show practically zero defect rates, the number of defective discharged asserrblies is quite high. This may require a well planned strategy for their disposal, and may require a different computation treatment for the source-term modeling of waste packages containing defected / leaker assemblies.

Pool-side fuel examinations and fission gas release data indicate that nearly 2,300 assemblies have failed or have damaged fuel rods (Gilbert et al.,1990). The total number of fuel rods with failed cladding is not known precisely. On.ne basis of visual observations and radiation monitoring of the pooi vater, it has been demonstrated that no significant degradation of Zircaloy cladding occurs during pool 4-15

m M

M M

M-M W --

Table 4-11. Spent fuel burnup and annual discharged quantities (adapted from U.S. DOE,1992b)

Average Burnup (mwd

kgU-')

Number of Anemblies kgU (x 10')

All Discharged Anemblics Equilibrium Cycle Discharges Year BWR PWR Total BWR PWR Total BWR PMR BWR PWR 1968 5

0 5

0.6 0.0 0.6 1.6 0.0 1.6 0.0 1969 96 0

96 9.8 0.0 9.8 15.2 0.0 15.2 0.0 1970 29 99 128 5.6 39.0 44.6 0.3 18.4 0.0 0.0 1971 408 113 521 64.0 44.5 103.5 5.8 23.9 16.6 0.0 1972 771 282 1.053 141.5 99.9 241.4 6.4 21.9 16.3 28.0 1973 577 165 742 95.2 67.1 162.3 12.4 23.7 15.9 28.1 p

1974 1,314 575 1,889 244.6 207.7 452.3 12.7 18.9 14.5 25.7 ch 1975 1,224 797 2,021 225.7 321.8 547.5 16.9 18.1 17.3 27.3 1976 1,663 931 2,594 297.1 401.0 698.1 13.4 22.2 16.7 25.8 1977 2.047 1,107

?,154 382.9 466.9 849.8 16.6 25.1 19.0 27.8 1978 2,239 1,666 3,905 383.2 699.0 1,082.3 19.8 26.4 22.1 28.7 1979 2,131 1,664 3,795 399.8 722.1 1,121.9 22.4 27.0 23.7 30.2 1980 3,330 1,456 4,786 619.9 618.1 1,237.9 22.4 29.7 23.3 30.7 1981 2,467 1,585 4,052 458.7 675.9 1,134.6 23.9 30.2 23.9 30.9 1982 1,951

!,493 3,444 357.2 641.3 998.5 24.7 29.7 24.9 32.3 4,478 491.3 773.1 1,264.5 26.7 30.0 26.7 31.6 1983 2,695 1,{0, 1984 2,736 1,951 4,687 498.0 847.6 1,345.7 25.4 29.4 25.4 32.0

M M

M M-M M

M M

M' d

~

Table 4-11. Spent fuel burnup and annual discharged quantitics (adapted from U.S. DOE,1992b) (cont'd)

Average Burnup (A1Wd

kgU-3) 3 Number of Assemblies kgU (x 10 )

All Discharged Assemblics Equilibrium Cycle Discharges Year BWR PWR Total BWR PWR Total BWR PWR 11WR PWR 1985 2,810 2,056 4,866 510.1 870.0 1,380.0 23.3 31.8 25.8 33.2 1986 2,551 2,348 4,899 458.2 1,022.5 1,480.6 21.0 30.4 27.3 33.7 1987 3,876 2,694 6,570 699.4 1,153.1 1,852.6 19.2 31.1 27.0 34.4 1988 2,957 2,731 5,688 535.8 1,162.5 1,698.2 24.1 33.0 27.0 35.3 1989 3,924 2,879 6,803 714.9 1,256.2 1,971.2 22.0 32.2 27.3 36.6 1990 3,486 3,594 7,080 633.0 1,552.2 2,185.2 25.0 33.8 27.8 35.8 Temps

3 139 142 0.5 45.7 46.2 0.0 27.0 27.0 1

45,293 32,105 77,398 8,227.0 13,687.3 21,914.3 21.3 29.7 24.2 32.4 "IUTAL

Temps are temporarily discharged assemblies, as of December 31,1990.

l m

m

=

m m

m e

'M M

M M

W W

M M

M-Table 4-12. Defective BWR fuels by assembly class and fuel design (Moore et al.,1990) ai I

GE BWlU2,3 GE BWR/4,5,6 (9 reactors)

(28 reactors)

Discharged Defective Percent Discharged Defective Percent Fuel Assembly Design Assemblics Assemblics Defective Assemblics Assemblics Defective GE Model 2 7 x 7 Fuel 6719 1469 21.9 1142 385 33.7 GE Model 3 Improved 7 x 7 Fuel 394 7

1.78 4936 130 2.63 GE Model 4 Original 8 x 8 Fuel 3876 1

0.03 3571 185 5.18 I

GE Model 5 8 x 8 Retrofit Fuel 792 1

0.13 3455 104 3.01 GE Prepressurized Fuel 1836 0

0.00 6591 144 2.18 GE Barrier Fuel 248 0

0.00 775 1

0.13 ANF 7 x 7 Fuel 260 0

0.00 Not Applicable ANF 8 x 8 Fuel 684 0

0.00 Not Yet Discharged GE - General Electric Company ANF - Advanced Nuclear Fuel

M M

M-4 Tahic 4-13. Defective PWR fuels by assembly class and fuel design (Moore et al.,1990)

WE 14 x 14 WE 15 x 15 WE 17 x 17 (6 reactors)

(10 reactors)

(33 reactors)

Fuel Assembly Discharged Defective Percent Discharged Ddettive Percent Discharged Defective Percent Design Assemhlics Auemblies Defective Assemblies Assemblics Defective Assemblics Assemblies ikfective WE Standard 592 1

0.2 1457 103 7.1 Not Applicable WE LOPAR 1409 77 5.5 3087 16 0.5 5102 99 1.9 WE OFA 88 1

1.1 266 1

0.4 628 1

0.2 WE VANTAGE 5 Not Yet Discharged Not Yet Discharged 4

0 0.0 ANF for WE 559 0

0.0 743 12 1.6 139 0

0.0 ANF TOPROD 299 1

0.3 Not Applicable Not Applicable WE - Westinghouse Electric Corporation ANF - Advanced Nuclear Fuel m

.m

+

i 1

l storage (Johnson,1979; Johnson,1977; Zima,1979; Bradley et al.,1981; Bailey and Tokar,1982).

However, conclusions about the likely performance of the cladding in a geological repository cannot be l

made from wet storage observations, since the conditions are quite different. On the other hand, I

conditions experienced in dry storage could yield information relevant to a repository situation because i

of similarities between the two environments. However, experience and quantitative data accumulated for the dry storage of spent fuel is limited.

j I

I I

I 4-20 I

-i 1

5 POST-DISCIIARGE SPENT FUEL DEGRADATION g

5.1 SPENT FUEL OXIDATION Upon breach of the waste package and failure of the fuel cladding, the speat fuel will be exposed to the repository environment-which during the early stages is expected to be dry or contain moist air at elevated temperatures ranging from 250 to 100*C. Under the anticipated conditions in an unsaturated repository, it can be expected, thermodynamically, that the UO fuel will oxidize completely 2

to UO. He sequence of oxidation is believed to be UO -U 0 -+U Or*U Og*UO *UO 2H O 3

3 4 9 3

3 3

3 2

(Einziger,1991). An uncertainty in modeling the long-term performance arises from the kinetics of the process.

The oxidation state of UO spent fuel after an extended residence in a moist air atmosphere at 2

elevated temperatures will affect the structural integrity of the fuel rod and the h. aching characteristics of the fuel pellets. As long as the fuel retains its cubic structure, the primary concern is the change in I

leaching rate as the O/M ratio increases. If low density phases such as U Og form, then additional g

phenomena must be considered: (i) release of a luge fraction of fuel matrix Kr and 14C inventory as gases (ii) spallation of fuel into powder, increasing the surface area available to a leachant, and (iii) stress from swelling fuel that is likely to split the cladding, thus decreasing the length of the flow path of the leachant to and from the fuel. The extent to which these effects may appear would depend on the oxide phases present in the fuel.

I Studies of spent fuel indicate that oxidation proceeds by forming U 0 along grain boundaries 4 9 and by subsequent growth of the U 0 into the individual UO grains (Woodley et al.,1989; Einziger 4 9 2

et al.,1991). He oxidation of spent fuel differs from unirradiated UO both in its rapid penetration 2

along grain boundaries and in the formation of a different oxidation product. An experime.:tal investigation reported in the literature used four LWR fuels that covered a range of burnup, grain size, and fission gas release, as shown in Table 5-1. He weight gains at 195*C due to oxidation in moist air are shown by the increase in the O/M ratio in Firure 5-1. Table 5-2 ranks the fuels by relative initial weight gain and lists some parameters, such as grain size, fission gas release, burnup, and fuel t3pe, that are commonly thought to affect oxidation rate. No simple correlation is apparent from these results.

The examination of spent fuels used in the investigation confinned that oxidation occurred by the formation and growth of U 0 along the original UO grain boundaries (Einziger et al.,1992). The 4 9 2

oxidation behavior observed in low gas-release fuels (ATM-105, ATM-104, and Turkey Point) consisted of uniform grain boundary oxidation extending from near the pellet outer edge to near the mid-radius.

Oxidized fuel particles from near the pellet centers in low-gas release fuels exhibited localized oxidation I

around the particle surfaces. In high-gas release fuel (ATM-106), uniform grain boundary oxidation occurred in samples which were from regions near the fuel pellet center. The reasons for this difference in oxidation are not understood, but are believed to be related to the fuel microstructure formed during j

burnup.

j 5.1.1 Fuel Condition Dependence The O/M ratio versus time data shown in Figure 5-1 vary initially by as much as a factor of four between fuels. However, the shape of these O/M ratio curves appear quite similar after a relatively I

short period of time. His is demonstrated in Figure 5-2a (Einziger et al.,1992) by the rate of O/M 5-1 I

I Table 5-1. Fuel specimen characteristics (adapted from Einziger et al.,1991)

I ATAf-104 ATAI-106 ATM-105 Turkey Calvert Calvert Properties / Parameters Cooper Point Cliffs Cliffs Fuel type BWR PWR PWR PWR 7x7 15 x 15 14 x 14 14 x 14 Nominal burnup, mwd

kgU-3 28 27 43 48 Fission gas release, %

0.6

< 0.3 1.1 18 I

NU 2.93 2.56 3.04 2.45 Initial enrichment, wt. %

Initial pellet density, % of theoretical 95 92 94 to 96 92 to 94 I

Postirradiation grain size, pm 11 to 15 20 to 30 10 to 13 7 to 15 Average nominal LHGR kW m-1 17 18.2 21 NA Postirradiation cooling time, yr 8.7 15.3 8.7 10.3 I

NA: Not available ATM: Approved Testing Material I

2.4 g

i R

u 4 3 2.3 O]

.9E=

'I

] 2.2 2

Sample ID O

I c.

N Turkey Point g

4 ATM-106

@ 2.1 A ATM-104 O

O ATM-105 O Turkey Point 8 ATM-103 I

I I

I 2.0 O

0.5 1.0 1.5 2.0 2.5 3.0 3

Oxidation Time. x 10 h Figure 5-I. O/M ratio with exidation time in flowing dry air at 195'C (Einziger et al.,1992) 5-2 I

.~

I Table 5-2. Ranking of fuels by weight gain (adapted from Einziger et al.,1992)

I Approximate Specimen Grain Size, Gas Release,

Burnup, Identification '

m Fuel Type mwd kgU-'I 104 5-15 1.1 PWR 43 I

105 11-15 0.6 BWR 28 103 18.5 0.3 PWR 30 106 10-13 18 PWR 48 I

TP 20-30

<0.3 PWR 27

Weight gain in descending order. Bracket indicates essentially equal weight gain.

I TP: Turkey Point fuel change, d(0/M)/dt, as a function of time. Although it is reported that after a short time (approximately 500 hours0.00579 days 0.139 hours 8.267196e-4 weeks 1.9025e-4 months at 195*C and 2000 hours0.0231 days 0.556 hours 0.00331 weeks 7.61e-4 months at 175*C) the d(0/M)/dt becomes essentially independent of fuel type, this conclusion is not supported by the data presented. However, it is clear from Fig. 5-2a that the I

rate of oxidation for all test samples reduces substantially with increasing time. Figure 5-2b shows the spent-fuel oxidation data for exposure up to 3000 hours0.0347 days 0.833 hours 0.00496 weeks 0.00114 months . From the graphs, it is clear that the rate of oxidation of the different spent-fuel samples test varies by as much as a factor of 10, even after 3000 hours0.0347 days 0.833 hours 0.00496 weeks 0.00114 months of oxidation. The oxidation curves also cross after approximately 600 hours0.00694 days 0.167 hours 9.920635e-4 weeks 2.283e-4 months of exposure.

I After a few thousand hours at 175*C, d(0/M)/dt is reported to be the same for BWR fuel of average grain size oxidized at a dewpoint of 80*C, and large-grained PWR fuel oxidized at dew points of -55 *C 5

and 80*C (factor of approximately 10 in moisture content). The observation is based on tests that have I

run for ~30,000 hours0 days 0 hours 0 weeks 0 months (Einziger et al.,1992). From this observation, one would conclude that the oxidation rate is independent of the moisture content. This fact could be of considerable significance in l

reducing the impact of some of the uncertainty related to possible changes in the moisture content of the l

unsaturated repository over the long period of regulatory interest.

5.1.2 Temperature Dependence I

l As shown in Figures 5-1 and 5-2, after a peak, d(0/M)/dt rapidly decreases with time, and the O/M ratio approaches a saturation plateau between 2.35 and 2.4 at 195'C. Longer term oxidation tests i

at 250 and 175*C indicate an O/M ratio value closer to 2.4 (Einziger and Strain,1986). U 0 is the 4 9u only observed phase at this O/M ratio. However, tests at higher temperatures indicate that this may be only a quasi-eo,uilibrium O/M ratio, and eventually oxidation to a higher state may occur. These observations are combined to generate a universal O/M ratio curve, as shown schematically in Figure 5-3.

l The data indicate that there is a minimum time given by 1.4+6, which is a function of temperature, 2

before detrimental lower density phases, such as U O, form (Einziger et al.,1992). Currently, it is not 3 g known why a plateau in the O/M ratio curve of Figure 5-3 occurs or if any mechanistic changes take

(

place in the fuel while it is on the plateau (Einziger et al.,1992). Since 6 was at least 9000 hours0.104 days 2.5 hours 0.0149 weeks 0.00342 months at 250*C for U 0g+, to transform to U Og, it is expected that the time for a higher phase such as U Og 4

3 3

to form at lower temperatures would be greatly extended, assuming an Arrhenius dependence.

5-3 lI

.~

I 1.0

[

LEGEND 0.8

-C}- ATM-104

-o-ATM-105

-M-Turkey Point i ATM-106 l 0.6 I

5 a

E 0.4 I

2 m

195'C Data 4

- I 0.0'/

0 2

4 6

3 Oxidation Tirre, x 10 h Hgure 5-2a. Time rate of change in O/M ratio as a function of oxidation time up to 5500 hours0.0637 days 1.528 hours 0.00909 weeks 0.00209 months (Einziger et al.,1992) 1.2 Sample ID i

1 E Turkey Point O ATM-106

^

B T

0.8 0 ATM-105 C

b 5

0.6 -

I

'R 0.4 I

195'C Data A

_- %Z g

0.0 I

I I

O 0.5 1.0 1.5 2.0 2.5 3.0 I

3 Oxidation Timo, x 10 h Figure 5-2b. Time rate of change in O/M ratio as a function of oxidation time I

up to 3000 hours0.0347 days 0.833 hours 0.00496 weeks 0.00114 months (Einziger et al.,1992) 5-4 I

I I

I E

9.

I

.9E m

g2.4 I

7

!=

5 5

g I

O I

'2.4 I

Time Figure 5-3. Schematic of O/M ratio in spent fuel as a function of oxidation time i

(Einziger et al.,1992)

The growth exhibits parabolic kinetics. Extrapolation of the oxidation rate indicates that the U 0 will remain stable for a period longer than 2000 years at 95'C (Einziger et al.,1992).

4 9 5.1.3 Spent Fuel Corrosion and Degradation l3 In a recently reported detailed investigation on the leaching behavior of spent UO fuel, test 2

j samples from both BWR and PWR fuels were used. The fuel / clad samples had a maximum burnup of approximately 40 mwd kgU-l (Forsyth and Werme,1992). As shown in Table 5-3, two aqueous lE leachants, namely, a synthetic groundwater (SGW) and DIW, were used. All tests were performed at

)g temperatures in the range of 20 to 25'C. The results of the investigation are summarized below.

Table 5-3. Composition (mmol L-1) of SGW at pH of 8.0-8.2 (Forsyth and Werme, 1992) 2 2

2 Na+

K+

Mg '

Ca +

Si HCO -

Cl-SO -

l 3

4 l

2.8 0.1 0.2 0.45 0.2 2.0 2.0 0.1 I

5-5 I

1 L

~

g..-

5.1.4 Oxic Conditions Oxic conditions are likely to be present, at least during the early part of the isolation period, I

in the geologic repository site currently being investigated in the United States. Therefore, understanding of the kinetics of release of actinides and Ession products under such conditions is of interest.

i 5.1.4.1 Release of Actinides The data on the concentration of U released in the two types of water are plotted against the I

cumulative contact time in Figure 5-4.

In the first week of contact with the SGW, the concentration 4 mol L-l and remains at that approximate level for 1000 days. At longer times, increases to 5 x 10 there appears to be a slowly increasing trend in concentration. For DIW, the U concentrations are lower, I

generally around I x 10 mol L-1 The Pu data are shown in Figure 5-5. The Pu concentrations after short-term exposures to SGW are relatively high, ranging from 1 x 10-9 to 5 x 10-8 mol L-1 At longer contact times, the concentrations decrease to values as low as 3 x 10-10 mol L-1 at 2000 days. However, the concentration is shown to increase to 1 x 10-9 mol L-1 after 2600 days, indicating possibly large scatter in data or perhaps a change in the mechanism leading to dissolution of i

precipitated Pu. In DIW, the concentration remains high, relatively independent of contact time, with I

a range in concentration of I x 10-8 to 1 x 10-7 mol L-1. The ratio between Pu and U normalized 1

releases is almost 10-2 for contact with groundwater at longer exposure times, while for DIW, the corresponding number is about 10. This is reponedly due to a combination of higher Pu concentrations I

and considerably lower U concentration in DIW as compared to groundwater, in which the complexation with carbonates results in an increase of several orders of magnitude in U solution concentrations.

l 5.1.4.2 Release of Fission Products Long-term performance assessment models should also account for fission products with shorter I

half-lives, such as Cs and Sr, to cover events such as early wa.,te package failures due to manufacturing defects or unanticipa;ed seismic events. Under oxic conditions, Cs, Sr, Tc, and I are fairly soluble in aqueous solutions and groundwaters. However, observed variation of their release behavior indicates that I

the release mechanisms could be different. Cs and I have been shown experimentally to migrate to grain boundaries, fuel surfaces, and fuel cladding as a result of in-pile service (MCC,1990). Further, at a higher linear power rating, the releases have been shown to correlate with release of the fission gases, such as Kr and Xe (MCC,1990). Therefore, it would be expected that in short-term leaching tests, the I

spent fuel will release these elements at a high rate. Figure 5-6 illustrates such a rapid initial release for 137 the Cs under oxic conditions for the three fuel samples used in the leaching tests.

9 90 The corresponding data for Sr shown in Figure 5-7 exhibit that, in short exposures, the Sr release rate is fairly constant, thereafter it decreases with time similar to 137Cs. Metallic inclusions of noble elements, such as Mo, Tc, Ru, Rh, and Pd, usually occur with a composition corresponding to their 99 i

respective fission yields. The release of Tc, which is usually present in abundance at or between the fuel grain boundaries in fission gas bubble sites, is believed to be controlled only by the oxidation of the inclusions. The inclusion sizes are in the nm to pm range (Thomas et al.,1989). Figure 5-8 shows99Tc release data under oxic conditions. The measured release rates for the two types of spent fuel are comparable within the wide scatter observed in the experimental data. The release rate in the range of 6 x 10-7 to 1 x 10-5 -l was maintained for a long time, leading to higher fractional release rates for d

We than those for 137 90 Cs and Sr after a few years (compare data in Figure 5-8 with those in Figures 5-6 and 5-7).

5-6 I

i i

1.0 E-03 3

Legend I

1.0 E-04 r

+ BWR SGW i

Q PWR SGW Y

A BWR DIW I

D 1.0 E-05 p

h

+

0 BWR DlW hi$2 0

+

a I

b Y

l 9 1.0 E-06 : --

e

-+

=

5 O

E A

i 1.0 E-07 A

A o

g 3

"5 1.0 E-08 :

A A

O A

i

'.5 g

1.0 E-09 2

3.0 3.5

,E O

0.5 1.0 1.5 2.0 3

Cumulative contact time, x 10 d Figure 5-4. Uranium concentration in deionized water (DRV) and synthetic groundwater (SGW) as a function of cumulative contact time under oxic conditions (Forsyth and Werme,1992) 1.0 E-06 5

Logend 4

5

+ BWR SGW I

O PWR SGW L 1.0 E-07 7 g

A gwa oiw O BWRDlW 2I

.0 A

A A

D A

i gs j 1.0 E-08

I i

+

8 o

- +$++D+0

+

E

+

I-

+

5 1.0 E-09

+

h 9

+

ji

+

1.0 E-10 I

O 0.5 1.0 1.5 2.0 2.5 3.0 3.5 8

Cumulative contact tirre, x 10 d Figure 5-5. Plutonium concentrations in DRV and SGW as a function of cumulative contact time I

under oxic conditions (Forsyth and Werme,1992) l 5-7 I

I 1.0 E-02 I-i ry LEGEND

.5 1.0 E-03 lg O Oskarshamn 1 BWR I

j 5

Eg C Ringhals 2 PWR C

+ Ringhals 1 BWR g5 I 1.0 E-04 r E

E I

tc

_!_ 1.0 E-05 Eh e

i a

I 2

1.0 E43 137gg o

I c

1.0 E-07

' ' '.0 O.001 0.01 0.1 1.0 10 3

Contact Time x 10 d I

Figure 5-6. Fractional release rates for 137Cs under oxic conditions (adapted from Forsyth and Werme,1992) i I

LEGEND i

5g O O Oskarshamn 1 BWR i

e

.0 E-05 c 1 O Ringhals 2 PWR i

O I

3

+

O O O Ringhals 1 BWR c =

O y 1.0 E-06 r s+e am

~

c O

g 1.0 E-07 7 n

i wsr o

u E

a 1.0 E-08 0.001 0.01 0.1 1.0 10.0 I

3 Contact Time x 10 d Figure 5-7. Fractional release rates forN I

Sr under oxic conditions (adapted from Forsyth and Werme,1992) 5-8 I

I i

1.0 E-03 E

LEGEND I

6 1.0 E-04 :

D Ringhals 2 PWR b

Q Ringhals 1 BWR e

I C

~

o 1.0 E-05 r a

2 D

O QD

[El j 1.0 E-06 Og N

i O

O D 2

I

.6 1.0 E-07 5

E NC tt I

1.0 E-08 O.001 0.01 0.1 1.0 10.0 3

Contact Trne, x 10 d I

Figure 5-8. Fractional release rates for "Tc under oxic conditions (Forsyth and Werme,1992) j 5.1.5 Reducing Conditions The effects of anoxic conditions on fractional release from a series of BWR fuel specimens are shown in Table 5-4. The change in redox condition has a major effect on the solution concentrations of 99Tc, decreasing it by 2 orders of magnitude, while only minor decrease in the cumulative release I

fractions were noted for 137 90 Cs and Sr. The average We concentrations under reducing conditions was found to be 6 x 10-9 mol L-1 This value should be compared with a Tc solubility of 3 x 10-8 mol L-1, which is controlled by TcO (a stable phase under experimental conditions).

I However, We was also present in the fuel in metallic form (which is not readily leachable). 'Therefore, 2

99 additional information would be required to make definitive conclusions about the release of Tc. For 137 90 Cs and Sr a comparison of their release rates gives more information on the mechanisms involved, as shown in Tables 5-5 and 5-6. The release of137Cs during the first 21 days of exposure is independent 90 of redox conditions, confirming that this represents rapid initial release. During the same period, the Sr release rate shows a more marked difference between oxic and anoxic conditions, particularly during the 4 d-l level. During the last second contact period, where the fractional release rate decreased to < 10 contact period, the rates for both nuclides are below this level, even under oxic conditions, but at this 13 stage Cs release is more redox-d endent. For the fiist time, its fractional release rate under reducing 137 conditions is lower than the rate for r, suggesting an attack at sites already relatively depleted in Cs, either during reactor operation or previous attack.

l 5.1.6 Comparison of Leaching Under Oxidizing Versus Reducing Environment In a study on the radionuclide release mechanisms of spent UO fuel, it is reported that U and 2

Pu releases are much smaller for both BWR and PWR fuels under reducing conditions than those under 5-9 I

~

g..-

I Table 5-4.

Cumulative release fractions in a 1-y test under oxic and anoxic conditions (adapted from Forsyth and Werme,1992)

Burnup Redox 01Wd kgU-2)

Condition Cs "Sr "Tc U

Pu 237 I

40.1 Oxic 7.8x10-3 7.6 x 10-4 1.1 x 10-3 9.1 x 10-3 8.7 x 10-6 41.4 Anoxic 7.0 x 10-3 2.1 x 10-4

< 2.1 x 10-3

< 2.1 x 10-6 ND 43.8 Oxic 8.1 x 10-3 7.5 x 10-'

1.7 x 10-3 9.5 x 10-5 6.2 x 10-6 44.9 Anoxic 7.4 x 10-3 1.7 x 10-8

< 2.3 x 10-3

< 2.2 x 10-6 ND 45.8 Oxic 8.1 x 10-3 7.1 x 10-4 7.3 x 10-3 6 0 x 10-5 8.4 x 10-6 ND = Not Detected 137 Table 5-5. Rates of fractional release of Cs under oxic and anoxic conditions (Forsyth and Werme,1992)

Contact (days)

Oxic Anoxic Oxic Anoxic Oxic 7

1.04 x 10-3 0.95 x 10--3 1.01 x 10-3 0.92 x 10-3 0.86 x 10-3 21 0.82 x 10-5 1.10 x 10-5 2.95 x 10-5 3.45 x 10-5 7.52 x 10-5 63 1.56 x 10-6 0.85 x 10-6 1.91 x 10-6 1.75 x 10-6 2.39 x10-6 91 1.13 x 10-6 0.23 x 10-6 1.41 x 10-6 0.46 x 10-6 1.39 x 10.-6 182 8.19 x 10-7 0.95 x 10-7 9.82 x 10-7 1.47 x 10-7 9.55 x10-7 Table 5-6. Rates of fractional release of MSr under oxic and anoxic conditions (Forsyth and I.

Werme,1992)

I Contact (days)

Oric Anoxic Oxic Anoxic Oxic 7

3.08 x 10-5 0.32 x 10-5 2.15 x 10-5 0.29 x 10-5 1.03 x 10-5 21 1.18 x 10-5 0.08 x 10-5 1.42 x 10-5 0.05 x 10-5 1.48 x 10-5 63 1.50 x 10-6 1.04 x 10-6 1.63 x 10-6 0.26 x 10-6 1.87 x 10-6

)

91 1.00 x 10-6 0.55 x 10-6 1.05 x 10-'

O.41 x 10-6 1.12 x 10-6 182 5.92 x 10-7 2.81 x10-7 1.05 x 10-7 0.41 x 10-7 5.03 x10-7 g

s.w I

oxidizing conditions (Werme and Forsyth,1988). A reduction of approximately 3 orders of magnitude is exhibited by U, and of spproximately 2 orders of magnitude for Pu in BWR fuel, while for PWR fuels, the trends are similar, although releases are smaller. Release of other elements such as Cs, Sr, Ba, and I

La are also at least an order of magnitude lower under reducing conditions than those under oxidizing conditions for BWR fuel, as shown in Figure 5-9. For PWR fuels, Sr exhibited a reduction in release of an order of magnitude under reducing conditions, while releases of Cs, Ba, and La were relatively I

unaffected by the change from oxidizing to reducing conditions, as shown in Figure 5-10.

5.1,7 Fuel Burnup Effects I

The effect of fuel burnup is shown in Figures 5-11 and 5-12, which present the cumulative 137 release fractions for Cs and 90Sr at various cumulative leachant contact times in oxic groundwater.

Figure 5-11 shows that the release fraction for 137Cs becomes practically time-independent after 28 days and varies along the fuel rod as related due to variations in burnup. The release shows an increase relative to lower burnup at a burnup of 35 mwd kgU~1 A similar effect is exhibited by the Sr 90 release, but the releases continue to increase with time. For exposure times longer than 28 days, the release fractions for both nuclides attained a maximum at a burnup of about 44 mwd kgU-1, but show an unexpected decrease at higher burnups. A similar pattern has been reported for Mc release, while U release was quite uniform along the axial fuel length. It is possible that the increase in release fractions for Cs, Sr, and Tc observed for fuel with a burnup of about 35 mwd kgU-l is related to the linear power of the fuel rod, as a result of migration and concentration of these radionuclides at the grain I

boundaries due to axial temperature gradient within the fuel rod. However, the rapid initial release of 137Cs shows a contact time dependence during the first 28 days, which is also a function of burnup. This suggests that leachant access to fuel surfaces can be at least a partial cause of the effect.

iR 5

5.1.8 Flowthrough Leaching Tests 4

E In an experimental study designed to determme the basic rate constants for the dissolution kinetics of UO and spent fuel, a single-pass flowthrough method was used (Gray et al.,1992). In such a

2 a system, U concentrations remain below solubility limits allowing relatively unambiguous reaction rates

g to be measured. Such measurements, together with the surface areas of the test specimens, can provide 3

relevant information for developing a kinetic model for UO dissolution. The statistically designed test 2

matrix covered four water chemistry variables: (i) temperature, 25 to 75*C, (ii) pH, 8 to 10, (iii) d M to 2 x 10-2M, and (iv) oxygen fugacity, 02 Partial i

carbonate / bicarbonate concentration,2 x 10 lg3 pressure 0.002 to 0.2 atm. The objective was to obtain data on UO: and spent fuel under equivalent controlled test conditions, in the hope of conducting a majority of the future tests on unirradiated UO2 I

The samples were fabricated from PWR fuel with a peak burnup of 33 mwd kgU-2 to reduce cost.

and a fission gas release of 0.25 percent.

Table 5-7 summarizes the dissolution conditions and test results. Figure 5-13 shows results I

typical of all ten runs. An early transient period, that usually lasted several days, was observed in each test, followed by stabilization of U dissolution rates to a nearly constant value independent of flow rate at all but the lowest flow rates. Most testing was performed at flow rates of 0.2 ml min-l. Figure 5-14 I

also shows dissolution rates for 137Cs, "Tc, and 90Sr. An example of the flow-rate-dependent release, shown in Figure 5-14, indicates that the U concentrations were directly proportional to the reciprocal of the flow rate except at the lowest now rates (reciprocal flow rates above 15 min mL-1).

I 5-11 I

r I

3.0 E-03 BWR Low Burnup Fuel 3.0 E44 -

C Oxid zing condnion E Reducing condnoon j 1.0 E44 --

3.0 E45 -

1 1.0 E45 -

4 er 3 o pas -

I 1.0 E46 -

3.0 E 07 -

1.0 E47 -

U Pu Cs Sr Ba La 1

Eierrent I

Egure 5-9. Comparison between fractional releases of uranium, plutonium, and fission products under oxidizing and reducing conditions in a low burnup BWR fuel (adapted from Werme and Forsyth,1988) 1.0 E-00 PWR Fuel 1.0 E Smuental Leaching L _j Oxideing condnion

]

E Reducing condnion I

I 1.0 E f 1.0 E 10 E $ 10 E I-1.0 E 1.0 E U Pu Cs

$r Ba La Element Egure 5-10. Comparison between fractional releases of uranium, plutonium, and fission products under oxidizing and reducing conditions in a PWR fuel (adapted

I from Verme and Forsyth,19SS) 5-12

.I r--*

c---,--

-r w

w

= -, - -

n

10

~

Legend I

n O

7 days S 8

- * - 28 days 7 # {'

x

-A-91 days

.l.

182 days i

J

';&,y'&

I

-8

- ---+--- 364 days

.y 1

y,

.5 v.

-6 1

u.

I 8,

~

,r's '

y

..+ -

c 4

%G7, R

I

-g 9

G5 2 I

O O

30 40 50 Burnup, mwd. kgU-1 137 Figure 5-11. Cumulative release fractions for Cs under oxic conditions for 7, 28,91,182, and 364-day leaching periods (Forsyth,1991)

Legend C

7 days I

S 8

-*- 28 days x

-A-91 days

~~,+.-- ----+., M,

,. #.. ~~,. -

d

- 182 days

b',/ -W

-'g",

+

---+-- 364 days 9

s # ^ * ~ ~ ~ ~ A.. '. 6 S 6 O

LL N

o V

_N

,g ',,, ", * ~ ~ ~ ~ g

,*/

e

+ *... + --.--+ /

A D

/

g h~..g I

/

N O

/ x/

MA' x

C 4 A.._._A-_..g' f

?

/

I

_.g Q

w

.x 5 2 0

I

~

C O

m 0

I 30 40 50 Bumup, mwd + kgU-1 Hgure 5-12. Cumulative release fractions for S'Sr under oxic conditions for 7, I

28,91,182, and 264-day leaching periods (Forsyth,1991)

I 5-13 I

M M

i Talle 5-7. Summary of spent fuel dissolution results (adapted from Gray et al.,1992)

U dissolution rate Measured Conc.of (mg m-2 d-3) 2-] + [IlCO -]

Temperature CO Temperature

[CO3 3

2 b

d No.

('C)

(molar)

corrected pil (ppm)

Days Avg.

2a '

1 50 2 x 10-3 8.7 to 8.9 117 16-37 6.34 0.55 2

50 2 x 10-3 8.7 to 8.9 117 16-37 7.05 1.44 3

50 2 x 10-3 8.7 to 8.9 117 16-37 5.07 1.35 4

23 2 x 10-2 8.0 to 8.2 11000 83-105, 116-130 3.45 0.51 5

75 2 x 10-2

'.7 to 9.9 55 38-95 11.6 2.3 6 I 74 2 x 10-4 7.3 to 7.9 125 22-81 8.41 1.33 Y

6s 75 2 x 10-4 7.3 to 7.9 125 22-75 8 79 1.74 7

23 2 x 10-4 9.7 to 10.2 0.9 37-78 0.413 0.138 8 h 23 2 x 10-2 9.I to 9.2 1030 22-59 2.83 0.30 9

23 2 x 10-3 9.9 to 10.2 7.7 5-32 2.04 0.56 Total carbonate concentration made up using appropriate amounts of Na2CO and NallCO.

3 3

b pil was measured at room temperature and a calculated temperature correction was applied, where necessary.

The water supply reservoirs were slowly and continuously sparged with gas containing these concentrations of CO, balance CO -free air, to 2

2 stabilize the pII.

d Time period over which the dissolution rate was averaged. This time period eliminates the transient period as well as periods when the flow rate was low enough to yield questionable results.

The standard deviation was calculated using the equation for a small number of observations. o = ((l'(x, -i)2 y f gy _ fjj u f

Same spent fuel specimen used in test 2.

8 Fresh spent fuel specimen used.

h Same spent fuel specimen used in test 7.

.A

~

k[O

i..

v IMAGE EVALUATION

's O

j/

N )/,7',/'#

g+, '

TEST TARGET (MT-3) f,f,:4%*

e,

/

\\

sh 8ge

~~

kY,p

'[$

ff' gg,

,gy 1.0 i22 l,l u,=, 2 Ri 1.8 pa:1 i' lL.25 !! l.4 Eij I.6 s

u-=

I 150mm y

4.- __

[

s Ng 4g;f

//&

s. ;A w

O e

zu v -

,N

/ gt,

r Ol l

'F j

Y 1

y

,f(g

'2' n'

,N,.-

O ab e..

IMAGE EVALUATION

///

a'A t

//>

TEST TARGET (MT-3)

'j$

g

,7 D'

4' I.o

- ?

l22 1._.g.

!?2.0 l,l Ws T!! l.8 hiita2+a

] l.25 ll 1.4 lll 1.6 i

n_=

m mm 4..

- 150mm l

_.._.._.___---------6" l

e, S 'k,

,\\

je s

k

$L.

syy,g~ })*

.q,,4,,4

+

O,,

,y~

y e

L-

x c$+

m x6 ' 4,

/.g g

f)0ihlji#p 5 @[ W)f)@

IMAGE EVALUATION

\\//g

/

t TEST TARGET (MT-3)

.v s

a i

l l

I ln

!:a e?a a 2.5 I. V fa, j'

=

(?! 2.2 u"

l I fli ijl 2.0 j

l*l

",_ t ;, I!w l 1.8

===

I l.25 1.4 i 1.6 I

L_

i I

i i

l 150mm l4 I*

6" l

e.

!E

/

m[P'I)S 4

//s s

x yf 4

t gg A /

y g p',K

,y-4 v g, / J

.~

4!L?6

,b d(

f r

y 7v

.s

,.~.., _ _ - _ _ - _ _ _._

~- -- -

' %A

' '10 0 O

, w

~

L.,

_n..._,.

x-7 sfs

,jj/,S y'e v hk = da f/s\\N\\\\\\

c.,

O tP IMAGE EVALUATION N&

\\/o/

the TEST TARGET (MT-3) f Q;![,

/-),*

/

Y)W'

&<e#

,, f

/////

4 I

l.0 i? %

u-n22

i. -

Licz u

'?

l,l i e.,=m lll 1.8 M== =

l 1.25 I

l.4 il I.6 1

I i l J

4--

150mm 4

4_____

6" f

,N'hp$

$"fll,/

chiV

<(&

,7,]/ @

s l4Rs l% l.i{f %,}f l

4R,[s/fr\\

/f g-w,p' Oy

+Q@

v e

t ge

~

1.

"I YaJ

IT *A E*U#

._..,_,.-x

~

I 12 12 OU 10 A 137Cs

- 10,b I

O STc

e D

pD Rate, mL min-1 E

E 5

3 $

5

~

2 I

0 O

O 20 40 60 80 100 120 140 160 Time, d Mgure 5-13. Dissolution rate of radionuclides from spent fuel grains in 2 x I

10-2M NaIICO/Na2CO solution at 20 to 25'C and pII 8.0 to S.2 (Gray et al.,

3 1992)

I 1400 1200 I

h 1000 CO 800 O

o

'5 600 N

d 400 200 I

O O

5 10 15 20 25 1/ Flow, min mL-'

I Mgure 5-14. Dependence of U concentration on reciprocal flow for spent fuel grains in 2 x 10-2M NalICO /Na2CO solution at 20 to 25'C and pII 8.0 to 8.2 3

3 (Gray et al.,1992) 5-15

m g..-

The analyses of the data yielded the activation energy of the dissolution process as 30 kJ mol-1 It is reportedly higher than a value of 20 kJ mol-1 obtained with unirradiated UO in 2

2 x 10-3 M NaHCO solution at pH 8.4 (Gray et al.,1992). Other 1" ature reported estimates are 19 3

to 25 kJ mol-1 between 25 and 150*C for spent fuel in groundwater containing approximately 10-3 M carbonate (Gray et al.,1992). Conclusions that can be drawn from the investigation, reported for 0.2 atm O partial pressure only, are that the spent fuel dissolution rate exhibits a weak dependence on I

2 carbonate / bicarbonate concentration, a moderate dependence on temperature, and is nearly independer.t of pH over the range tested, 8 to 10.

l 5.1.9 Unsaturated Repository Simulating Tests In the unsaturated repository site being investigated at Yucca Mountain, Nevada, only a limited amount of water is expected to contact the fuel under anticipated conditions. Under these circumstances, laboratory tests on spent fuel samples totally immersed in large amounts of water may only represent conditions in which flooding of the repository occurs due to water intrusions through flaws in the rock.

In order to evaluate the performance of spent fuel, tests representative of other conditions for the unsaturated repository environment need to be used in spent fuel testing. In one such experimental investigation, the reaction between water and simulated wasteform under potential repository conditions I

was examined (Wronkiewicz et al.,1992). The method involved periodically dripping small amounts of water on samples that were maintained in temperature-regulated stainless steel test vessels. After contacting the sample, the injected. vater was allowed to migrate down the sides of the.anple, drip from I

the bottom, and collect in the test vessel. The test thus simulates an environment where the waste container develops breaches along its top and bottort, allowing free flow of water through the container and limited contact time with the waste form. Sample fuel pellets were fabricated from pressed and I

sintered UO powder with the natural isotopic abundance of U. The pellets were clad in a Zircaloy-4 2

tubing with bare UO surfaces exposed on the top and bottom of the tubing. The leachant for the 2

experiments was tuff-core equilibrated groundwater obtained from a well (J-13) in the vicinity of the proposed Yucca Mountain repository site. The congosition of the equilibrated water (EJ-13 water) is I

presented in Table 5-8. Tests were conducted at a nominal temperature of 90*C.

Table 5-a. Chemical composition (mg L-1) of EJ-13 water including carbon <ontaining species, anions, and cations (pH = 8.2, Solution U = 0.0024)

(adapted from Wronkiewicz et al.,19921 TOCa TIC b IICO -

C O/-

Cl-F-

NO -

2 2

3 6.9 18.4

< 0.5

<0.5 8.6 2.8

<0.5 8.2 SO/-

3 2

2 d

B+

lj +

Ca +

Mg +

g+

Na+

Si +

21 0.16 0.044 9.08 0.96 8.08 46.5 34.4 I

= Total organic carbon b

= Total inorganic carbon I

5-16 I

The results show that secondary uranyl alteration minerals began to form, and notable increases in U release occurred after the first 13 to 20 weeks of testing. This change was interpreted to reflect the onset of oxidative dissolution of the UO surface and the formation of dehydrated schoepite. After I

approximately 1 year, samples displayed a sharp increase in U release. This rapid pulsed release is the 2

most conspicuous feature of the entire 4.5-year reaction sequence. Sample surfaces at this stage were partially covered by dehydrated schoepite crystals, as well as minor schoepite, soddyite, and Fe-oxides.

The UO samples also displayed extensively corroded grain boundaries and a dusting of fine dislodged 2

surface granules. These loosened particles were believed to be responsible for the rapid release of U noted during this interval.

I After 1.5 to 2.0 years of reaction, the rapid pulsed release of U subsides, and release rates declined to an apparent steady-state level that continued up to 4.5 years. On average, only 5 mole percent of the total U released from the samples remains in solution or as suspended fine particulate matter. The remainder either settled to the bottom of the test vessel or was reprecipitated on the surfaces of test components. 'Ihe top sample surfaces were found to be covered by uranyl silicate phases I

(uranophane, boltwoodite, and sklodowskite) that had incorporated cations from the El-13 water, as well as U from the dissolution of the UO surfaces and/or previously deposited minerals. The similar 2

mineralogy observed between the unsaturated tests, shown in Figure 5-15, and those of natural uraninite occurrences exposed to surfacial weathering suggests that these experiments have formed a long-term alteration sequence of uran), silicates that may be expected to control U release in the potential repository. Both the dissolution of the UO: surface and the quantity ofloosened fine-grained particulates on the sample surface appeared to have declined while uranyl silicates were precipitating.

'Ibe authors of the investigation conclude that although experiments conducted with UO may not be entirely representative of reactions expected from spent fuel, the experiments may provide insights into the potenti.11 behavior of spent fuel dissolution since spent fuel is composed dominantly of UO 2 (Wronkiewicz et al.,1992). The experiments indicate that U releasm in a moist oxidizing environment will be controlled by a complex interaction of a number of processes including dissolution rates of grain i

I boundaries, spallation of surface particles, secondary phase formation and stability, and mode of contact and composition of water with the waste form. Many of the more soluble radionuclides that migrate to the grain boundary regions during reactor operations (e.g.,137Cs and 90Sr) are readily mobilized during I

saturated dissolution tests with spent fuel (Johnson and Shoesmith,1988). The rapid disintegration of grain boundaries, noted in the reported experiment, indicates that soluble grain nuclides segregated at grain boundaries may also be readily mobilized in an unsaturated repository environment. Furthermore, the spallation of UO granules from the sample surfaces may also continuously expose new grain 2

boundaries to corrosive fluids, thereby prolonging the release period for grain boundary radionuclides.

The resultant pulsed release of U associated with spallation of UO granules has implications for the 2

release of radionuclides from the UO matrix of spent fuel. Since the amount of U released in the 2

experiments appears to be largely controlled by the migration of the fine-grained particulate matter, it follows that the migration of other radionuclides contained in the UO: matrix of spent fuel (Pu + Am) may be controlled by particulate dispersion rather than by solubility of the UO matrix or the individual 2

radionuclides alone. Migration of these radionuclide may now be determined by the ability of the repository hydrologic system to transport fine solids away from the waste package environment as colloidal matter. The findings of this investigation provide additional challenges to the source-term model development exercise.

5-17 I

3 Paragenetic Sequence Time, years 1

2 3

4 Ongt of U Dissolution i

Pulsed Releasa I

U Oxide Hydrates Soddyite g

j U Alkali Silicates f

f 1

?

I O

50 100 150 200 250 Time, weeks t

I Figure 5-15. Interpretive paragenetic sequence formed on top surfaces of altered uraninite pellets. Onset of U dissolution and pulsed-release periods determined from solution data. (Wronkiewicz et al.,1992)

I I

I 5-18 I

j, 6 DISCUSSION OF SPENT FUEL LONG-TERM PERFORMANCE ISSUES Source-term models are expected to provide estimates of the release of radionuclides from the waste packages under geologic repository conditions. Estimates are required for conditions that are likely to prevail for the 10,000 years of regulatory concern and for additional scenarios which could be anticipated.

The reliability of the source-term model will depend on how appropriately the understanding of the quan" ties and characteristics of the spent fuel, waste package materials, repository environment (such as 1

I geochemistry, temperature, radiolysis products), and temporal changes in all these aspects over the 10,000 yeais after the repository closure are included in the model. Uncertainties in spent-fuel behavior could undoubtedly influence the evaluation (or assessment) of performance in the repository. The output of the 1

source-term modeling exercise can be simply stated as the expression of the ' release rates and quantities' I

of the radionuclides from the spent fuel as a function of time over 10,000 years. The calculated releases r

will be compared with the allowable releases per 10 CFR 60 tojudge if the repository design will provide i

the required level of isolation of the spent fuel. There is little uncertainty in the total inventory of the radionuclides in the spent fuel, and the task at hand would be to model release rates and cumulative releases during the 10,000-year period. A similar argument would apply to the case of vitrified (glass) wasteforms.

I Based on the discussion of the inventories of the radionuclides in the cladding and spent fuel, the characteristics c# the fuel and cladding (i.e., inhomogeneous distribution of the radionuclides), the oxidation behavior of the fuel, and its leaching response when in contact with moist and aqueous environments, it can be concluded that releases of radioactivity in a geologi,,epository will be greatly influenced by the processes considered. For an unsaturated repository, there could be extended periods of time during which the fuel may not come in contact with water, but the characteristics of the fuel may change due to oxidation in air, leading to increased surface area and release of radioactive gases in the l

fuel matrix. Subsequent exposure of this ' altered

spent fuel to aqueous leaching would be quite different t

I from the event in which solid UO in the pellet form comes in contact with water. The time at which the 2

waste package fails is also critical, as it will have a bearing on the type and the amounts of radiolysis products present in the groundwaters that will contact spent fuel. Radiolysis products have a direct

, I influence on the pH of the water, which is one of the major parameters that influence spent fuel dissolution characteristics. At the present time, only a few source-term models exist. Although a detailed review of the individual models is outside the scope of this report, a survey of the general capabilities I

and limitations approaches used for the development of the current source-term models indicate that some significant conclusions drawn from experimental data and analyses of the spent fuel and cladding are absent or inadequately addressed in current source-term models. Some performance factors that could I

modify or increase the calculated performance of the spent fuel over 10,000 years, are listed below. The literature provides a reasonable amount of data and technical information in these area, except where noted, for their inclusion in current models.

j

Processes over 10,000 years of repository performance. Source-term models need to include the degradation behavior of the cladding and spent fuel during the period of exposure to high temperature (above the boiling point of water). Extended exposure at such temperatures I

could substantially alter the fuel characteristics, which will influence the release of radionuclides upon subsequent exposure to liquid phase groundwater. The degradation of fuel, which may be called ' alteration', will include oxidation of the fuel to higher oxidation I

states, substantiJ increase in the surface area, which will lead to increases in the calculated 6-1 I

.c release rates per unit volume of fuel, and rapid release of fission gases normally trapped in the spent UO matrix.

2 e Use of data related to leaching behavior of spent fuel with ' altered

characteristics in modeling releases. The majority of the literature reported data are on release of radionuclides from spent UO fuel rather than fuel with higher oxygen to uranium ratios, 2

such as U Og The models should also address the scenario where a mix of unaltered spent 3

fuel, semi-altered (quasi-equilibrium state discussed in the report), and fully altered spent fuel are releasing radionuclides in the near-field of the. repository.

Data related to leaching of spent fuel in repository specific geochemistry should be incorporated in the mechanisms which are incorporated in the models. The geochemistry should incorporate the effects of radiolysis products, leading to changes in the water pH, redox conditions, and presence of waste package corrosion products. Since the heat output of the containers with vitrified wastes is lower than containers with spent fuel, it is conceivable that these containers will corrode / breach before spent fuel containers because contact with liquid water will occur at shorter times. Under such a scenario, the geochemistry of the groundwaters will also be modified by the releases from the vitrified I

high-level wastes. The large number of containers with vitrified wastes (about one-third of the anticipated 40,000 to 50,000 containers) indicates that the groundwaters could contain a substantial amount of radionuclide releases from vitrified wasteforms.

Although most source-term models acknowledge the radioactivity that might be released in Id the form of gases, this aspect has generally been addressed by C releases from the crud / corrosion products on the cladding and the spent fuel. This approach will result in underestimation of the contribution of the gaseous releases from the cladding and fuel, as less than 1 percent of the wall thickness of the cladding is oxidized. An extended period of exposure of the cladding to high temperature can oxidize the cladding leading to additional I

release. Similarly, as stated in an earlier item, further oxidation of spent fuel in the repository will break up the fuel pellets and generate particulates leading to release of gaseous and volatile fission products to the near-fleid environment. Although the volume of the cladding is estimated to be less than 5 percent of the fuel, it contains approximately 50 percent of 'C (the other 50 percent being in the fuel).

e Currently it is estimated that at least 1 out of every 10,000 fuel rods will be in the breached condition at the time of disposal. Although releases of radionuclides from fuel assemblies with such breached rods can be detected while in service in-reactor, individual rods are difficult to identify or remove economically. As such, it is likely that they will be disposed of along with the unbreached rods. Their distribution among the different waste packages can be known with certainty. Based on the number of containers likely to be emplaced in the first repository, one can assume 4,000 to 5,000 breached fuel rods. The radionuclide 1

release characteristics of these rods will be different from the unbreached rods under most

{

repository scenarios. The contribution of the breached rods to the source term will be considerable, and needs to be accounted for. Current models assume no breached rods.

)

e According to the SCP design, it is estimated that there will be approximately 1,000 fuel rods per waste package. The DOE in assuming 'no-credit

for the cladding assumes that upon failure of a given waste package, fuel contained in all 1,000 rods will be exposed to the I

I

n repository environment. This may neither be a realistic situatian nor even a conservative one. For instance, if the failure of the waste package is due to flooding of the repository leading to corrosion of the metallic container, we are assuming that spent UO fuel contained I

in approximately 1,000 rods inside the waste package is exposed simultaneously to 2

groundwater. Under such a scenario, the source-term model will use leaching mechanisms and experimental data related to leaching of ' unaltered

spent fuel to calculate release of radionuclides. This scenario is inconsistent with a distribution of failure of the cladding with time, which would lead to alteration of the spent fuel, e.g., substantially increased surface area per unit volume, different release mechanism, rapid release of fission gases, etc.

I Therefore, the source-term models need to incorporate a distribution of cladding failure as a function of time.

I-

Most current source-term models are based on the assumption that maximum release of a particular radionuclide in the leachant (groundwater) will be limited (controlled) by the solubility limit of that radionuclide. This assumption would be valid provided 100 percent I

of the release of all radionuclides were in solution in the aqueous phase. There is evidence that such is not always the case. A number of elements are likely to occur in the form of colloids and precipitates in a geologic repository. Since the precipitates and colloids arc not I

in solution, they provide an add-on term to the source-term models. Although, the fraction of releases of actinides known to form of colloids and precipitates in not known precisely at present, the assumption used in some elementary treatments in source-term models of a maximum of 30 percent could be a gross underestimation.

The presence of organic material and microbes in water is known to influence the migration of radionuclides in underground environments.

Microbial activities expected in an I

unsaturated repository may provide a significant add-on term to the source-term modeling and needs to be investigated. Since the studies related to microbial activities and their influence on the alteration of spent-fuel characteristics are practically non-existent, additional I

experimental investigations in this area are warranted to estimate the contribution of the i

potential microbial activities to the source-term.

The EBS components are expected to be constructed from a number of materials. Most studies related to the failure of the waste package concentrated on the corrosion failure of the metallic container as a result of groundwater / container interaction, and do not address electrochemical effects between the fuel and the cladding and the fuel and the container material. Insufficient consideration of potential galvanic effects in selecting waste package materials could adversely affect the fuel cladding failure distribution as a function of time and could also accelerate the spent fuel degradation / dissolution. Therefore, electrochemical interactions between the waste package materials, spent fuel cladding, and the spent fuel need to be evaluated and possibly incorporated in the source-term models.

I I

6-3 j

I

c I

7 REFERENCES g

Adachi, T., T. Muromura, H. Takeishi, and T. Yamamoto.1988. Metallic phases precipitated in UO 3

fuel: II. Insoluble residue in simulated fuel. Journal of Nuclear Materials 160: 81-87.

Andrews, M.G., H.R. Freeburn, and W.D. Wohlsen.1979. Performance of combustion engineering fuel I

in operating PWRs. American Nuclear Society Topical Meeting-LER Fuel Performance. La Grange Park, IL: American Nuclear Society (ANS).11-19.

Bailey, W.J., A.B. Johnson, Jr, and R.W. Lamben.1986. Integrity of spent fuel and pool components during wet storage. Transactions of the American Nuclear Society. 50: 116.

Bailey, W.J., and A.B. Johnson, Jr.1990. The expected condition of spent fuel on delivery to a repository. High Level Radioactive Waste Management. La Grange Park, IL: ANS. 1437-1443.

Bailey, W.J., and M. Tokar.1982. Fuel Performance Annual Reportfor 1981. Washington, DC: U.S.

Nuclear Regulatory Commission (NRC).

Baily, W.E., J.S. Armijo, J. Jacobson, and R.A. Proebstle.1979. BWR fuel performance. American Nuclear Society Topical Meeting-L%R Fuel Performance. La Grange Park, IL: ANS.1-10.

I Barner, J.O.1984. Giaracterization of LMR Spent Fuel MCC-Approved Testing Material-ATM-101.

PNL-5109. Richland, WA: Pacific Nonhwest Laboratories (PNL).

I Bell, M.J.1973. ORIGEN. 7he ORNL isotope generation and depletion code. ORNI 428. Oak Ridge, TN: Oak Ridge National Laboratory (ORNL).

I Bennett, M.J., J.B. Price, and P. Wood.1987. Influence of manufacturing route and burnup on the oxidation and fission gas release behavior of irradiated uranium dioxide in air at 175-400"C.

Proceedings of the International Workshop on 01emical Reactivity of Oxide Fuel and Fission Product Release. Harwell, UK: United Kingdom Atomic Energy Agency (UKAEA).

Bowman, S.M.1992. PC input processor for ORIGEN-S. Proceedings of the 7hird International Conference on High Level Radioactive Waste Management. La Grange Park, IL: ANS. 88-92.

Bradley, R.E., W.J. Bailey, A.B. Johnson, Jr., and L.M. Lowry.1981. Examination ofzircaloy-clad spentfuel after extended pool storage. PNL-3921. Richland, WA: PNL.

Brodda, B.G., and E.R. Merz.1984. The leaching behavior of zircaloy cladding and its impact on the conditioning of waste for final disposal. Nuclear Technology. 65: 432-437.

Croff, A.G.1983. ORIGEN-2: A versatile computer code for calculating the nuclide compositions and characteristics of nuclear materials. Nuclear Technology. 62: 335-352.

Cubicciotti, D., J.S. Sanecki, R.W. Strain, S. Greenberg, L.A. Neimark, and C.E. Johnson.1976. The nature offission product deposits inside LWR Fuel Rods. SRI Project No. 4197. Menlo Park, CA: SRI International.

l l

7-1 I

J

3..-

Einziger, R.E., S.C. Marschman, and H.C. Buchanan.1991. Spent-Fuel Dry-Bath oxidation testing.

Nuclear Technology. 94: 383-393.

Einziger, R.E., S.D. Atkin, D.E. Stellrecht, and V. Pasupathi.1982. High temperature postirradiation materials performance of spent PWR fuel rods under dry storage conditions. Nuclear Technology. 57: 65-80.

I Einziger, R.E.

1991. Effects of an oxidizing atmosphere in a spent fuel packaging facility.

FOCUS'91-Conference on Nuclear Waste Packaging. La Grange, Park, IL: ANS.

Einziger, R.E., and R.V. Strain.1986. Behavior of breached PWR spent-fuel rods in an air atmosphere between 250*C and 360*C. Nuclear Technology. 75: 82-95.

EPR1.1989. Fuel consolidation demonstration: Spentfuel storage area analysis. EPRI NP-6551. Palo Alto, CA: Electric Power Research Institute (EPRI).

EPRI.1984. Surmillance ofLWR spentfuel in wet storage. EPRI-NP-3765. Palo Alto, CA: EPRI.

EPRI.1986. Fuel and pool component performance in storage pools. EPRI-NP-4651. Palo Alto, CA:

EPRI.

I Fish, R.L., and R.E. Einziger.1981. A perspective onjission gas releasefrom spentfuel rods during geologic disposal. HEDL-TME 81-3. Richland, WA: Hanford Engineering Development Laboratory (HEDL).

Forsyth, R.S.1991. Spent fuel degradation. Scientific Basis of Nuclear Waste Management XIV.

Pittsburgh, PA: Materials Research Society (MRS). 212: 177-188.

Forsyth, R.S., and L.O. Werme.1992. Spent fuel corrosion and dissolution. Journal of Nuclear Materials 190: 3-19.

Garzarolli, F., R. von Jan, and H. Stehle.1979. Main causes of fuel element failure in water-cooled power reactors. Atomic Energy Review. 17: 31-128.

Gilbert, E.R., W.J. Bailey, A.B. Johnson, Jr., and M.A. McKinnon.1990. Advances in technology for storing

WR spent fuel. Nuclear Technology. 89: 141-161.

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

Guenther, R.J.1983. Results of simulated abnormal heating events forfull-length nuclearfuel rods.

PNL-4555. Richland, WA: PNL.

I Guenther, R.J., D.E. Blahnik, T.K. Campbell, U.P. Jenquin, J.E. Mendel, L.E. Thomas, and C.K.

Thornhill, 1988. 01aracterir.ation of spent f.tel approved testing material ATM 103.

PNL-5109-103. Richland, WA: PNL.

7-2 I

S Guenther, R.J., D.E. Blahnik, T.K. Campbell, U.P. Jenquin, J.E. Mendel, L.E. Thomas, and C.K.

Thornhill. 1991. Otaracterization of spent fuel approved testing material AIM 105.

PNL-5109-105. Richland, WA: PNL.

Guenther, R.J., and 3.O. Barner.1981. Performance ofAnnular-Coated, Annular, Vipac, and Reference t

Test Rods During Steady-State Irradiation in HBnR IFA-518.1.

PN1s3952 (DOE /ET/34215-23): Richland, WA: PNL.

Guenther, R.J., E.R. Gilbert, S.C. State, W.L. Partain, J.R. Divine, and D.K. Kreid.1984. Assessment ofdegradation concernsfor spentfuel, high few! wastes, and transuranic wastes in monitored i

retrievable storage. PML-4879. Richland, WA: PNL.

I Guenther, R.J., D.E. Blahnik, L.E. Thomas, D.L. Baldwin, and J.E. Mendel.1990. Radionuclide distribution in LWR spent fuel. Proceedings ofSPECTRUM'90 - Nuclear and Ha:ardous Waste Management International Topical Meeting. La Grange Park, IL: ANS. 120-123.

Heeb, C.M., M.A. McKinnon, and J.M. Creer.1990. ORIGEN-2 predictions of spent fuel decay heat rates and isotopic content compared with calorimeter and key actinide measurements.

Proceedings of SPECTRUM'90-Nuclear and Hazardous Waste Management International Topical Meeting. La Grange Park, IL: ANS. 116-119.

i Hirabayashi, T., T. Sato, C. Sagawa, N.M. Masaki, M. Saeki, and T. Adachi.1990. Distributions of radionuclides on and in spent nuclear fuel claddings of PWRs. Journal of Nuclear Materials 174: 45-52.

I Johnson, A.B., Jr.1977. Behavior of spent nuclear fuel in water pool storage. BNWL-2256. Richland, WA: Battelle Pacific Northwest Laboratories (BPNL).

Johnson, A.B., Jr.1979. Spent fuel storage experience. Nuclear Technology. 43: 165-173.

Johnson, A.B., Jr., E.R. Gilben, and R.J. Guenther.1982. Beharlor ofspent nuclearfuel and storage I

system components in dry interim storage. PNL-4189. Richland, WA: PNL.

Johnson, A.B., Jr., and E.R. Gilbert.1983. Technical basisfor storage of zircaloy-cladfuel in inen gases. PNL-4835. Richland, WA: PNL.

Johnson, E.R.1986. Centralized disassembly and packaging of spent fuel in the DOE spent fuel management system. Tr mactions of the American Nuclear Society. 52: 73-74.

Johnson, L.H., and D.W. Sh< esmith.1988. Spent fuel. Spent Fuel, Radioactiw Waste Formsfor the I

Future. W. Lutze at.d R.C. Ewing, ed. North-Holland Physics Publishing, Amsterdam, ' Die Netherlands. 635-693.

Johnson, L.H., D.W. Shoesmith, G.E. Lunansky, M.G. Bailey, and P.R. Tremaine.1982. The mechanisms of leaching and dissolution for UO f

b uel pellets under simulated disposal conditions: Nuclear Technology. 56: 238-253.

I 7-3

m I.

Katayama, Y.B., D.J. Bradley, and C.O. Harvey.1980. Status report on LWR spentfuelIAEA leach rests. PNL-3173. Richland, WA: Phl.

Klein, J.A.1987. The integrated database program: An executive database of spent fuel and radioactive l

waste inventories, and characteristics. Proceedings of the WASTE MANAGEMENT '87 l

Symposium. Tucson, AZ: University of Arizona. 483-488.

' I Kleykamp, H.1979. The chemical state of LWR high-power rods under irradiation. Journal ofNuclear Materials 84: 109-117.

locke, D.H.1975. Review of experience with water reactor fuels - 1968-1973. Nuclear Engineering and Design. 33: 94-124.

Manaktala, H.K.1990. Potential degradation modes of spent fuel zirconium alloy cladding in s geological repository emitonment. Ninth International Symposium on Zirconium in the Nuclear Industry. Kobe, Japan. American Society for Testing and Materials (ASTM).

Manaktala, H.K. 1991. Degradation Modes of Nuclear Reactor Fuel Cladding. Paper No. 91037.

CORROSION /91. Houston, TX: National Association of Corrosion Engineers (NACE).

Moore, R.S., K.J. Notz, and C.G. Lawson.1990. Classification of LWR defective fuel data.

I Proceedings of the SPECTRUM'90 Nuclear and Hazardous Waste Management International Topical Meeting. La Grange Park, IL: ANS. 129-132.

Moore R.S., and K.J. Notz,.1991. Tools for LWR spent fuel characterization: Assembly classes and fuel I

designs. Proceedings ofthe Second AnnualInternational Conference on High Lewl Radioactive Waste Management. La Grange Park, IL: ANS.1: 91-95.

Notz, K.J., R. Salamon, T.D. Welch, R.J. Reich, and R.S. Moore.1992. Spent fuel characteristics provided by the CDB - An Update. Proceedings ofthe ThirdInternational Conference on High Level Radioactiw Waste Management. La Grange Park, IL: ANS. 122-130.

Ocken, H.1982. Fission gas release from fuel rods irradiated in U.S. commercial LWR. American Nuclear Society Topical Meeting on LWR Euended Burnup-Fuel Performance and Utilization.

La Grange Park, IL: ANS. 3/59-3/75.

Olander, D.R.1976. Fundamental aspects of nuclear reactorfuel elements. TID-26711-Pl. Technical Information Center. Energy Research and Development Administration. Oak Ridge, TN:

ORNL.

Parks, C.V.1992. Overview of ORIGEN-2 and ORIGEN-S: Capabilities and Limitations. Proceedings ofthe ThirdInternational Confererce on High Lew! Radioactive Waste Management. La Grange Park, IL: ANS. 57-64.

Pearcy, E.C., and H. K. Manaktala.1992. Occurrence of metallic phases in spent nuclear fuel:

Significance for source term predictions for high level waste disposal. Proceedings ofthe 7hird I

7-4

International Conference on High Lewl Radioactive Waste Management. La Grange Park, IL:

ANS. 131-136.

PNL.1990. Materials Characterization Center (MCC) Burnup and Fission Gas Release Distribution Workshop. Seattle, WA. June 27-28,1990. Richland, WA: PNL.

Thomas, L.E., R.E. Einziger, and R.E. Woodley.1989. Microstructural Examination of Oxidized Spent PWR Fuel by Transmission Electron Microscopy. Journal ofNuclear Materials 166: 243-251.

Tsoulfanidis, N., and R.G. Cochran.1991. Radioactive waste management. Nuclear Technology. 93:

263-304.

I U.S. DOE.1992a. Garacteristics ofPotentialRepository Wastes. DOEIRW-0184-RI. Washington, DC:

U.S. Department of Energy (DOE).

U.S. DOE.1992b. Spent Nuclear Fuel Dischargedfrom U.S. Reactors 1990. SRICNEAFl92-01.

U.S. DOE.1982. Spent Fuel and Radioactive Waste Inventories, Projections, and Garacteristics.

DOE /NE-0017-1.

U.3. DOE.1988a. Garacteristics ofSpent Fuel, High-Izvel Waste, and OtherRc?oactive Wastes Which May Require long-Term Isolation. DOE /RW-0184. Oak Ridge, TN: ORNL. Vol.14.

U.S. DOE 1988b. Garacteristics of Spent Fuel, High-Level Waste, and Other Radioactive Wastes Which May Require Long-Term Isolation. DOEIRW-0184. Oak Ridge, TN: ORNL. Vol. 7-8.

U.S. NRC.1990. Title 10 U.S. Code of Federal Regulations Part 60. Disposal of High-Level Radioactive Wastes in Geological Repositories. Ofce ofthe Federal Register: Washington, DC:

I NRC.

VanLuik, A.E., M.J. Apted, W.J. Bailey, J.H. Haberman, J.S. Shade, R.E. Guenther,.R.J. Serne, E.R.

Gilbert, R. Peters, and R.E. Williford.1987. Spent fuel as a waste form for a geologic disposal: Assessment and recommendation on data and modeling needs. PNL-6329. Richland, WA: PNL.

I Welch, T.D., K.J. Notz, and R..J. Andermann, Jr.1990. Important parameters in ORIGEN2 calculations of spent fuel compositions. Proceedings of SPECTRUM'90 Nuclear and Hxardous Waste Management International Topical Meeting. La Grange Park, IL: ANS. 133-136.

Welch, T.D., and K.J. Notz,1990. Utilization of ORIGEN-2 by the characteristics database. Proceedings ofSPECTRUM '90-Nuclear and Hacardous Waste Management International Topical Meeting:

La Grange Park, IL: ANS. 72-76.

Werme, L.O., and R.S. Forsyth.1988. Spent UO fuel corrosion in water: Release mechanisms.

2 Scientific Basis of Nuclear Waste Management XI. Philadelphia, PA: MRS. 112: 443-452.

I 7-5

.I. '

Williamson, R., and S.A. Beetham.1987. Fission product and U Og particulate emission arising from 3

the oxidation of irradiated UO : Preliminary studies. Proceedings ofthe International Workshop 2

on Chemical Reactivity of Oxide Fuel and Fission Product Release. Berkeley, U.K.

I l

Wood, P., M.J. Bennett, M.R. Houlton, and J.B. Price.1985. Oxidation and fission gas release behavior of irradiated uranium dioxide in air bciow 400*C. Proceedings of the BNES Conference on I

Nuclear Fuel Performance. Stratiord-on-Avon, U.K.

i Woodley, R.E.1983. The characteristics of spent LWR fuel relevant to its storage in geologic Repositories. HEDL-TME 83-28. Richland, WA: HEDL.

Woodley, R.E., R.E. Einziger, and H.C. Buchanan.1989. Measurement of the oxidation of spent fuel between 140*C and 225'C. Nuclear Technology. 85: 74-88.

Wronkiewicz, D.J., J.K. Bates, T.J. Gerding, E. Veleckis, and B.S. Tani.1992. Uranium release and secondary phase formation daring unsaturated testing of UO at 90*C. Journal of Nuclear 2

Materials 190: 107-127.

Wuertz, R., and M. Ellinger.1985. Source term for the actinide release from a repository for spent LWR fuel. Scientific Basis of Nuclear Waste Management IX. Philadelphia, PA: MRS. 50:

393-400.

Yang, R.L., and D.R. Olander.1981. Behavior of metallic inclusions in uranium dioxide. Nuclear Technology. 54: 223-233.

Zacha, N.J.1988. Fuel rod consolidation. Nuclear News. 31:(3) 68-73.

Zima, G.E.,1979. An Evaluation of Potential Gemical/ Mechanical Degradation Processes Afecting I

FuelandStructuralMaterialsunderLong-Term WaterStorage.NUREGICR-0668. Washington, DC: NRC.

I I

I l

u 1

ML20056D854

Text

Dear Mr. Shelor:

Enclosure:

Navigation menu

Search