UO2 compatibility in liquid metal bonded light water reactor fuel elements

MISSING IMAGE

Material Information

Title:
UO2 compatibility in liquid metal bonded light water reactor fuel elements
Physical Description:
ix, 76 leaves : ill. ; 29 cm.
Language:
English
Creator:
Dubecky, Mark Andrew, 1970-
Publication Date:

Subjects

Subjects / Keywords:
Materials Science and Engineering thesis, M.S
Dissertations, Academic -- Materials Science and Engineering -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (M.S.)--University of Florida, 1995.
Bibliography:
Includes bibliographical references (leaves 74-75).
Statement of Responsibility:
by Mark Andrew Dubecky.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002045657
oclc - 33399217
notis - AKN3586
System ID:
AA00009021:00001


This item is only available as the following downloads:


Full Text












UO2 COMPATIBILITY IN LIQUID METAL BONDED LIGHT WATER
REACTOR FUEL ELEMENTS

















BY

MARK ANDREW DUBECKY


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE



UNIVERSITY OF FLORIDA


1995















ACKNOWLEDGEMENTS


I would like to express my appreciation and gratitude to

Dr. Richard Connell, Jr., my committee chairman, who has

taught me so much about the engineering profession during my

undergraduate and graduate studies. In addition, I would like

to thank Professor James S. Tulenko, Dr. Robert T. DeHoff, and

Dr. Paul H. Holloway. Members of my committee, these

individuals offered their time and wisdom while providing me

with guidance through my graduate education. I would like to

thank Thad Adams for passing the reins patiently and answering

any questions I may have had about the equipment or other.

Also deserving of gratitude is Professor Emeritus Glen J.

Schoessow, who taught me anything I wanted to know about

nuclear reactors and also served as my mechanical consultant.

I would like to thank the Office of Energy Research at

the Department of Energy and especially the Materials Science

and Engineering Department for supporting my research and

education. I would also like to thank B&W Nuclear

Technologies and A.E.C.L. Laboratories for their material

donations.

A special thanks to my wife Kelly is in order for her

support of my decision to continue my education.















TABLE OF CONTENTS


ACKNOWLEDGMENTS ...........................................ii

LIST OF TABLES..............................................v

LIST OF FIGURES...........................................vi

ABSTRACT........................ .................... .... viii

CHAPTERS

1 INTRODUCTION............................................ 1
1.1 Background......................................... 1
1.2 LBLWR Fuel Design................................. 1
1.3 Advantages of LBLWR Fuel Design.................... 3
1.4 Liquid Metal Concerns............................6
1.5 Liquid Metal Corrosion............................. 7
1.6 Previous Work on LBLWR Fuel....................... 8
1.7 Related Work............. .... .....................10
1.8 Radiation Effects on Materials ................... 11
1.8.1 Radiation Damage Mechanisms.................. 11
1.8.2 Radiation Effects on Ceramics................ 13
1.8.3 Radiation Effects in Liquid Metals........... 13
1.8.4 Radiation Effects on Corrosion............... 14
1.9 Scope of Research................................ 14

2 EXPERIMENTAL PROCEDURES................................. 16
2.1 Materials...................... ................ 16
2.2 Sample Preparation................................ 19
2.3 Testing Times and Temperatures................... 22
2.4 Metallographic Preparation....................... 25
2.5 Characterization Methods........................ 25
2.6 Average Loss of Wall Thickness Measurements......27

3 RESULTS AND DISCUSSION..................................29
3.1 Introduction..................................... 29
3.2 Intermetallic Compound Formation................. 29
3.2.1 Pb:Sn:Bi SOC Results......................... 30
3.2.2 Pb:Sn:Bi LOCA Results......................... 35
3.2.3 Sn:Bi:Ga Results.............................41
3.3 Consumption of Zircaloy.......................... 45
3.3.1 Intermetallic Layer Formation................ 45
3.3.2 Loss of Tube Wall Thickness Results..........49
3.3.3 Zircaloy-4 Hardness After Exposure........... 53
3.4 Liquid Metal Properties .......................... 56
3.5 U02 Properties ................................ 56









3.6 LBLWR Compatibility in a Reactor Environment.....63
3.6.1 SIMFUEL Results............................... 63
3.6.2 Compatibility Changes Under Radiation........ 64

4 CONCLUSIONS.............................................. 67
4.1 Introduction ..................................... 67
4.2 Results of Materials Compatibility for Liquid
Metal Bonded Light Water Reactor Fuel Elements...67
4.2.1 Loss of Tube Wall Thickness Results.......... 67
4.2.2 Intermetallic Compound Formation............. 68
4.2.3 Results of Hardness Testing.................. 69
4.2.4 UO2 Compatibility in the Pb:Sn:Bi LBLWR
Fuel ....................................... 70
4.3 Prediction of Compatibility in a Reactor
Environment ...................................... 70
4.4 Topic for Future Investigations .................. 70

APPENDIX A: PREPARATION OF SIMFUEL........................ 72

APPENDIX B: METALLOGRAPHIC PREPARATION .................... 73

LIST OF REFERENCES ........................................ 74

BIOGRAPHICAL SKETCH ....................................... 76















LIST OF TABLES


Table Page

2.1 ASTM B350 chemical composition specifications
for reactor grade zircaloy-4 ....................... 16

2.2 Composition of 6 at% burnup percent SIMFUEL........17

2.3 Grade and minimum composition of the components
used to produce the Pb:Sn:Bi alloy................. 18

2.4 Grade and minimum composition of the components
used to produce the Sn:Bi:Ga alloy..................18

3.1 Material constants of microstructural constituents.38

3.2 Enthalpies of Formation of oxides observed in
Pb:Sn:Bi LBLWR fuel samples ........................61















LIST OF FIGURES
Figure Page

1.1 Schematic showing section of current LWR fuel
design including thermal resistances................ 2

1.2 Schematic showing section of proposed LBLWR
fuel design......................................... 4

2.1 Schematic showing the location of the notch placed
into U02 pellets prior to testing ................. 20

2.2 Schematic showing the preparation of specimens
for furnace testing................................. 21

2.3 Barnstead-Thermolyne tube furnace apparatus........23

2.4 Schematic showing the position of the specimen
within the furnace apparatus ....................... 24

3.1 Backscattered electron micrograph of Pb:Sn:Bi
500 hour SOC sample showing ZrSn. massive
intermetallic particles.(l,200x) ...................31

3.2 Zirconium-Tin Binary Phase Diagram (taken from
ASM Handbook on Alloy Phase Diagrams).............. 34

3.3 Backscattered electron micrograph of Pb:Sn:Bi
5,000 hour SOC sample.(200x) .......................33

3.4 Optical photomicrograph and electron microprobe
results of Pb:Sn:Bi 24 hour LOCA sample.(400x).....37

3.5 Backscattered electron micrograph of Pb:Sn:Bi 24
hour LOCA sample showing intermetallic layers
corresponding to phases in Zr-Sn binary phase
diagram.(750X) ..................................... 39

3.6 Optical photomicrograph of Pb:Sn:Bi 24 hour LOCA
sample showing microhardness indentations in the
(A) massive intermetallic particle (l,000x) and
(B) intermetallic layer.(500x)..................... 40

3.7 Optical photomicrographs of (A) Sn:Bi:Ga and (B)
Pb:Sn:Bi 500 hour SOC samples.(100x) ...............43

3.8 Optical photomicrographs of Sn:Bi:Ga 24 hour
LOCA sample.(200x) .................................44









3.9 Backscattered electron micrograph of Sn:Bi:Ga 24
hour LOCA sample showing U-Sn-Ga compounds at
the top of saw cut:(A) (700x), (B) (2,000x)........ 45

3.10 Optical photomicrograph of Pb:Sn:Bi 24 hour LOCA
sample showing jagged ZrSn2 intermetallic
layer.(1,000x) ..................................... 47

3.11 Optical photomicrograph of Pb:Sn:Bi 24 hour LOCA
sample showing change in intermetallic layer at
base of saw-cut.(200x) ............................. 49

3.12 Optical photomicrograph of Pb:Sn:Bi 24 hour LOCA
sample showing change in tube wall thickness at
base of saw-cut.(100x).................. ........... 51

3.13 Loss of tube wall thickness results of Pb:Sn:Bi
SOC samples............................ ............. 52

3.14 Loss of tube wall thickness results of Pb:Sn:Bi
LOCA samples....................................... 53

3.15 Vickers microhardness results of the Pb:Sn:Bi
LBLWR fuel components in the SOC samples as a
function of testing time............................ 55

3.16 Vickers microhardness results of the Pb:Sn:Bi
LBLWR fuel components in the LOCA samples as a
function of testing time........................... 56

3.17 Backscattered electron micrograph of a Pb:Sn:Bi
24 hour LOCA sample showing zirconium oxide layer
present at the top of saw-cut:(A) (220x),
(B) (700x).............................. ........... 59

3.18 Backscattered electron micrograph of a Pb:Sn:Bi
5,000 hour SOC sample showing zirconium oxide phase
present in the saw-cut.(4,000x)......................60

3.19 Backscattered electron micrograph of Pb:Sn:Bi 24
hour LOCA sample showing growth of ZrSn2 MI
in intimate contact with U02.(1,500x) .............. 63

3.20 Backscattered electron micrograph of Pb:Sn:Bi
1,000 hour SOC SIMFUEL sample.(360x)...............66

3.21 Backscattered electron micrograph of Pb:Sn:Bi
1,000 hour SOC SIMFUEL sample showing zirconium
oxide layer within saw-cut.(1,000x)................67


vii















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science


U02 COMPATIBILITY IN LIQUID METAL BONDED LIGHT WATER
REACTOR FUEL ELEMENTS

By

Mark Andrew Dubecky

May, 1995




Chairman: Dr. Richard G. Connell, Jr.
Major Department: Materials Science and Engineering

An innovative design that places a liquid metal alloy in

the diametral gap between UO2 fuel pellets and zircaloy-4

cladding of light water reactor (LWR) fuel rods has been

proposed. This design improves thermal conductivity from the

centerline of the fuel rod to the coolant allowing for lower

operating temperatures, lengthened fuel life, and enhanced

safety in loss of coolant accidents (LOCA). Recent studies

have demonstrated acceptable compatibility between a ternary

lead-tin-bismuth alloy (33 wt% each) and zircaloy-4.

Materials compatibility among the three fuel rod components--

UO2, zircaloy-4, and liquid metal alloy--is reported.

SIMFUEL, a simulated high-burnup UO2 based nuclear fuel, has

also been studied in place of UO2 to determine if some adverse

reaction occurs among the liquid metal and simulated fission


viii









products. Elevated temperature tests have been made at 750F

and 1500F to simulate Standard Operating Conditions (SOC) and

LOCA conditions, respectively. Intermetallic compound

formation was characterized by optical and electron microscopy

as well as by electron probe x-ray microanalysis.

Microhardness tests were performed on the pellet, clad, and

solidified alloy to determine if any variations in the

mechanical properties of the system had taken place over time

as a result of interactions with the liquid alloy.

Temperature and liquid metal volume were observed to effect

the morphology and growth of the Zr-based intermetallics.















CHAPTER 1
INTRODUCTION


1.1 Background


The Department of Energy project "An Innovative Fuel

Design Concept for Improved Light Water Reactor (LWR)

Performance and Safety" has been conducted at the University

of Florida as an interdisciplinary effort between the

Departments of Nuclear Engineering Sciences and Materials

Science and Engineering. The project goal was to develop a

fuel design which yields improvements in both safety and

thermal/mechanical performance. From the outset of the

research, it was proposed that a Liquid metal Bonded Light

Water Reactor (LBLWR) fuel concept serve as the basis for

investigation.


1.2 LBLWR Fuel Design


Current Light Water Reactor (LWR) fuel assemblies consist

of U02 pellets encased in zircaloy-4 cladding. To facilitate

the loading of pellets during fabrication, the clad ID is

designed larger than the diameter of the solid pellet. Prior

to sealing each fuel assembly, a backfill of helium gas is

delivered to improve heat transfer from the fuel pellets to

the coolant. A schematic of this design is shown in Figure

1.1. Also shown in this figure are the percentages of
































Figure 1.1


COOLANT
7.3%




ZIRCALOY
4.7%


Schematic showing section of current LWR fuel
design including thermal resistances.


UO2 GAS
53% GAP
35%







3

thermal resistance for heat transfer from the centerline of

the fuel to the coolant. At the beginning of life, thirty-

five percent of the thermal resistance of the fuel occurs in

the gas gap (1). It is in this area that the LBLWR fuel

design attempts to improve. The proposed idea, shown in

Figure 1.2, replaces helium with a low melting alloy, liquid

at operating temperatures, to provide higher thermal

conductivity and improved heat transfer.




1.3 Advantages of LBLWR Fuel Design


The LBLWR fuel concept under development offers reduced

centerline operating temperatures and reduced stored heat in

the fuel. Computer modeling of LBLWR fuel was performed using

a modified version of the nuclear ESCORE (EPRI Steady-State

Core Reload Evaluator Code) fuel performance code (2). The

modified code, ESBOND, has predicted a decrease of 3500F in

centerline temperature (1).

The amount of heat stored in the fuel directly affects

the temperatures reached during a Loss Of Coolant Accident

(LOCA). By reducing the heat stored, the fuel temperatures

reached are significantly decreased. This, in turn, lowers

the risk posed by the high-temperature zirconium-water

reaction. Effects of this reaction include loss of cladding

integrity and production of explosive hydrogen gas. These

phenomenon could lead to a reactor core melt down or hydrogen

explosion. By reducing the temperatures reached during a



















COOLANT





LIQUID ZIRCALOY
METAL


Figure 1. 2


Schematic showing section of proposed LBLWR
fuel design.


U02







5

LOCA, the LBLWR fuel provides a larger margin for error,

slowing the rate at which the high-temperature zirconium-water

reaction occurs. Upon a 500F decrease in clad temperature,

the rate of this reaction is reduced by almost two orders of

magnitude (3).

The following additional advantages accrue as a result of

temperature reduction:


1. A reduced Doppler Effect and increased Doppler

temperature coefficient.

2. A lower rate and reduced amount of fission gas

release.

3. A lower rate of U02 swelling.


The Doppler Effect leads to the broadening of high

neutron cross section energies or resonances as the

temperature increases (4). This phenomenon effectively

reduces the reactivity and aids in the shutdown of the reactor

during a transient. Thus, by reducing the fuel temperature,

the shutdown margin of the fuel is increased.

The Doppler temperature coefficient(a) directly relates

the change in reactivity(p) to changes in fuel temperature(T)

by the following equation:

Ap = aAT

This coefficient is temperature dependent, increasing with

decreasing fuel temperature as the case in the LBLWR fuel

design. This has important implications when increasing the

power output of the reactor since smaller changes in

reactivity produce larger changes in the fuel temperature.







6

Reducing the rate and amount of fission gas produced in

the fuel lowers the risk of overpressurization and allows

fuels to be used to higher burnups. Initially, a gas pressure

of approximately 450 psi. exists in the fuel rods; however,

this amount increases during operation, eventually reaching a

pressure of 2,200 psi. at the end of life. Prior to this

time, the internal rod pressure is below that of the

surrounding coolant water (2,200 psi) and creepdown of the

clad occurs until it meets the expanding pellet. The time at

which the pellet and cladding meet in the LBLWR fuel is

lengthened by approximately 100 days over current fuel as a

result of the slower rate of U02 swelling. This reduces the

chance of failures by pellet-cladding interactions (PCI) and

leads to improved fuel-clad interaction performance

characteristics (5).

The net result of these advantages is the possibility of

increasing the life of fuel elements to higher burnups. A

higher degree of burnup has positive environmental

implications since the amount of waste generated is reduced

for a given amount of electricity produced. Thus, less

irradiated nuclear fuel will require storage.



1.4 Liquid Metal Concerns


The compatibility of liquid metal, zircaloy-4, and UO2 is

of considerable importance for this application. Any forms of

liquid metal corrosion which may jeopardize the structural









integrity of the zircaloy cladding must be addressed.

Cladding failures can result in the loss of fuel material and

spread of contamination.



1.5 Liquid Metal Corrosion

Two forms of liquid metal corrosion, dissolution and

alloying (intermetallic formation), exist in the LBLWR fuels.

Direct dissolution of a solid metal occurs when solid surface

atoms dissolve into a liquid metal (6). This continues to

occur until a solubility limit is reached in the liquid metal.

The solubility limit can be predicted by a phase diagram of

the solid-liquid system.

As is expected, the amount and rate of dissolution is

strongly dependent upon the operating temperature and volume

of liquid metal present. The rate of dissolution also is time

dependent on the ratio of liquid metal volume to solid-liquid

interfacial area. Consider two systems having the same solid-

liquid metal couple and solid surface area, but having

different liquid metal volumes. Initially, the rate is the

same in both systems no matter what ratio applies. After a

finite period of time, the system with the smaller volume of

liquid will have a concentration closer to its solubility

limit. The driving force for further dissolution in this

system is less than the system containing the larger volume of

liquid metal. Thus, the larger the liquid metal

volume/interfacial area, ie. the higher this ratio, the

greater is the amount of solid metal dissolution.







8

In direct alloying, the interaction between liquid and

solid results in the formation of surface films or typical

diffusion layers of intermetallic compounds and solid

solutions. These layers, if strongly bonded to the surface,

may serve as a barrier to diffusion and reduce the rate of

reaction between the solid and liquid metal (7). Also,

depending on the mechanical properties of these layers, they

may enhance or worsen the structural integrity of the corroded

member.



1.6 Previous Work on LBLWR Fuel


The materials compatibility of zircaloy-4 and two liquid

metal alloys was investigated by Thad Adams at the University

of Florida prior to this research (8). He studied the

compatibility of the eutectic lead-bismuth alloy and a lead-

tin-bismuth alloy (33-33-33 wt%) in zircaloy-4 cladding alone,

as well as with A1203 pellets to simulate a reactor fuel rod.

His experiments were conducted at temperatures predicted by

the ESBOND code to simulate both Standard Operating Conditions

(SOC) and Loss Of Coolant Accident (LOCA) conditions. Adams'

SOC samples were tested at 750F for time periods up to 3,500

hours. The LOCA tests were conducted for time periods up to

24 hours at temperatures from 1,200-1,5000F. Radioactive

species, originating in the fuel during SOC, continue to decay

and produce heat in the absence of the fission process. A

testing time of 24 hours was chosen to be certain that the

LBLWR fuel can withstand these decay heats.







9

Adams analyzed his polished samples in cross-section with

optical microscopy to characterize the loss of tube wall

thickness as a function of exposure time (8). He concluded

that the Pb:Sn:Bi alloy showed better compatibility than the

eutectic Pb:Bi alloy. His results predicted a loss of less

than 1% over 5,000 hours at SOC for the Pb:Sn:Bi alloy.

Adams observed the formation of an intermetallic layer on

surface of the zircaloy in samples containing both alloys (8).

Because the ternary alloy showed better compatibility with the

zircaloy-4, more detailed studies were confined to exposures

with that alloy. He analyzed the intermetallic layer using

electron microscopy and electron probe x-ray microanalysis and

determined the intermetallic layer to be the compound ZrSn2 in

the Pb:Sn:Bi samples. Further examination of his Pb:Sn:Bi

samples containing A1203 pellets revealed the presence of the

Massive Intermetallic (MI) particles discussed later (3.2).

Adams anodized several of his samples to look for another

form of liquid metal attack not previously mentioned (8).

Intergranular penetration/liquid metal embrittlement results

from preferential attack on the grain boundaries of a solid by

a liquid metal. His results showed no evidence of

intergranular penetration by the liquid metals into the

zircaloy. This result is consistent with the literature which

claims liquid metal embrittlement does not occur in the

presence of intermetallic compound formation (9).










1.7 Related Work


Several investigations on the dissolution of UO in

molten zircaloy have been performed at AECL Research operating

at Chalk River Laboratories in Ontario, Canada (Conversation

with P.G. Lucuta, AECL Research, March, 1995). In a recent

study, crucibles machined from SIMFUEL pellets(a UO2 based

fuel discussed in 2.1) were filled with zircaloy and heated

rapidly in argon to 2,000 or 2,2000C for 15 or 30 minutes.

The specimens were sectioned and analyzed using electron

microscopy, Energy Dispersive Spectroscopy (EDS), and

Inductively Coupled Plasma (ICP) emission spectrometry.

Analysis of the specimens revealed the following three

distinct macroscopic regions: a central melt region, a

transition zone, and the residual crucible. The central melt

region of the 2,000C specimens contained dendrites of three

intermixed phases within a Zr-rich alloy matrix. The three

intermixed phases consisted of a ceramic (U,Zr)O2-x phase, a

ZrO2-type phase, and an intermetallic Zr-Sn(O) phase of an

approximate composition corresponding to Zr5Sn3007. The

transition zone consisted of the following phases which varied

in abundance from the crucible side to the melt region: the

ceramic (U,Zr)02- phase, a (Zr,U)O2- phase containing traces of

fission products, and the O-saturated Zr matrix phase. No

significant phases developed in the residual crucibles;

however, the extent of metal precipitation was increased. The

2,2000C samples showed no additional phases. Knowledge of the







11

existence of these phases may be useful in describing the

results of this research.




1.8 Radiation Effects on Materials


While this research did not involve studies of samples

exposed to a reactor environment, it is important to

understand the effects of radiation on each component of the

LBLWR fuel.


1.8.1 Radiation Damage Mechanisms


In order to predict the compatibility of the LBLWR fuel

in a reactor situation, one must have an understanding of how

radiation interacts with materials. Radiation can be

classified as either energetic particles or electromagnetic

radiation (10). The energetic particles which include

fission fragments and fast neutrons are of primary concern in

solid materials. Electromagnetic radiation consists of gamma

and bremsstrahlung radiation and typically results in

ionization and electron excitation. Electromagnetic radiation

is of little importance in metals, but is important in non-

metals such as plastics, elastomers, and ionic or covalent

bonded materials.

The different characteristics associated with each of the

energetic particles govern the damage they can produce.

Fission fragments are large and only travel distances of a few

microns. As a result, the effects of fission fragments are







12

usually limited to fuel-bearing materials. Fast neutrons, on

the other hand, are relatively small with high energies. They

are capable of traveling long distances and producing a large

amount of damage when interacting with surrounding atoms.

Energy loss of energetic particles occurs (1) by elastic

collision with lattice atoms, and (2) by excitation and

ionization through charge interaction (10). While an

energetic particle loses its energy in one of two ways, the

damage created is thought to occur in several different ways.

Schottky or Frenkel defects are created by elastic collisions

that result in the ejection of an atom from its lattice site.

In addition to these displacement phenomenon, "spike" effects

are believed to exist. These effects involve the coordinated

behavior of many atoms.

A thermal spike occurs when an energetic particle glances

off a stationary atom. The collision induces an oscillation

which is transferred to neighboring atoms. The vibrational

period of 1013 to 10-14 seconds produces a violent agitation in

the vicinity. "This energy vibrationall) creates a momentary

region of high temperature that, in a metal, may involve

thousands of atoms and result in a temperature of perhaps

1,000K for probably less than 10-10 sec (10)." The spot

heating and rapid quenching associated with a thermal spike

may assist in activated processes such as phase changes,

diffusion, disordering, annealing, etc. It has also been

suggested that this sudden heating or even melting could

produce dislocations through strain and distortion of the

surrounding lattice.







13

An even more drastic event, known as a displacement

spike, has been predicted as a fast-moving atom rapidly

decelerates due to collisions with other stationary atoms. A

primary knock-on is a fast-moving atom created from a

collision with a fast neutron. Once the energy of the

displaced atom falls below a certain transition value, its

mean free path between collisions becomes approximately atomic

spacing. At this point, the atom is rapidly brought to rest

among an intense shower of secondary displacements. As a

result, 10,000 or more atoms are violently raised to the

molten state with turbulent flow. Complete rearrangement,

rapid quenching, and resolidification in alignment with the

surrounding lattice occur over a very short period of time.

Vacancies and interstitials are believed to be annealed in

this region with the damage showing as dislocation loops and

regions of misorientation.


1.8.2 Radiation Effects on Ceramics


Ceramics are relatively resistant to radiation damage.

A swelling effect is observed in many ceramics, but most of

the changes that occur are related to the electronic

properties. Properties such as optical absorption, magnetic

susceptibility, paramagnetic resonance, and electrical

conductivity have been known to change as a result of both

particle bombardment and electromagnetic radiation.










1.8.3 Radiation Effects in Liquid Metals


Except for developing radioactivity, liquid metals appear

to be unaffected by radiation. Interactions between energetic

particles and liquid metal atoms such as the ones described

above theoretically should occur. The movement of liquid

metal atoms, however, is already so high that no effect on the

liquid metal attack due to these collisions is expected.


1.8.4 Radiation Effects on Corrosion


The effect of radiation on corrosion could play an

important role in the LBLWR fuel design. The literature

suggests that corrosion by liquid metals is seldom affected by

radiation. In the LBLWR fuel, however, enhanced corrosion due

to radiation may occur. Fast neutrons and other energetic

particles tend to disrupt protective films subjecting freshly

exposed metal to further attack (10). This phenomenon could

occur to the intermetallic layers which form in the LBLWR fuel

as was discussed in 1.5.



1.9 Scope of Research


The compatibility among liquid metals, UO2, and zircaloy-

4 was investigated for direct application to the liquid metal

bonded light water reactor fuel design. The primary liquid

metal alloy studied was lead-tin-bismuth (33-33-33 wt%);

however, an alloy of tin-bismuth-gallium (Sn:Bi:Ga 48-48-4







15

wt%) was also the basis of an abbreviated investigation. In

addition, the compatibility among zircaloy-4, Pb:Sn:Bi, and

SIMFUEL, a simulated high-burnup UO2-based nuclear fuel, was

investigated (11).

The main objectives of this research were as follows:

1) To characterize the materials compatibility of the

liquid metal bonded fuels using optical and electron

microscopy, electron probe x-ray microanalysis (microprobe),

and Vickers microhardness testing.

2) To predict the compatibility of the liquid metal

bonded fuels in an actual reactor environment with the aid of

studies involving SIMFUEL. It is important to note that no

laboratory studies involved exposure in a reactor environment.

3) To suggest topics for future investigation.















CHAPTER 2
EXPERIMENTAL PROCEDURES


2.1 Materials


The zircaloy-4 cladding and natural U02 pellets used in

this investigation were supplied by B&W Nuclear Technologies

of Lyncburgh, Virginia. The composition of the zircaloy

follows ASTM standard B350 and is given in Table 2.1.


Table 2.1 ASTM B350 chemical composition specifications for
reactor grade zircaloy-4.

ELEMENT COMPOSITION RANGE Wt%
Sn 1.20-1.50
Fe 0.18-0.24
Cr 0.07-0.30
Fe+Cr 0.28
0 0.10-0.15
C 0.010-0.018
Si 0.007-0.012
Zr balance


The zircaloy clad donated consisted of both 15x15 and

17x17 sized tubes. The 15x15 size clad has an OD of

approximately 0.430 inches with a 0.026 inch wall thickness.

The OD of the 17x17 sized tube is approximately 0.380 inches

with a wall thickness of 0.024 inches. The natural UO2

pellets sized for a 15x15 tube have a diameter and length of

0.370 and 0.408 inches, respectively. Natural UO2 consists of

0.7% Um5 (the fissile isotope) and approximately 99.3% Un8.







17

Eight 17x17 sized SIMFUEL pellets were donated by

A.E.C.L. Research of Ontario, Canada. SIMFUEL, a simulated

high-burnup UO2-based nuclear fuel, contains rare earth and

other oxides as well as spherical metallic precipitates which

exist in nuclear fuels at the end of life (11). The

composition of SIMFUEL pellets corresponding to 6 at% burnup

is given in Table 2.2.


Table 2.2 Composition of 6 at% burnup percent SIMFUEL.

COMPOUND* ORIGEN Codeb WDXc UO, Matrixd
U02(Pu) 95.31 -- -
BaCO3 0.311 0.26 0.02
CeO2(Np) 0.526 0.58 0.56
La203(Am,Cm) 0.194 0.20 0.19
MoO3 0.730 0.62 --
SrO 0.110 0.13 0.06
Y203 0.061 0.06 0.06
ZrO2 0.601 0.60 0.40
RhO3 0.034 0.04 --
PdO 0.440 0.42 0.01
RuO2(Tc) 0.764 0.71 --
Nd203(Pr,Pm,Sm) 0.912 0.85 0.85

Concentrations increased to account for elements in parenthesis.
Used to calculate the fission-product compositions (60 kW/m).
Average of five different WDX scans in 200 square micron areas.
Average of five different WDX scans in UO, matrix (spot mode).


This table lists the composition of irradiated fuel as

predicted by the ORIGEN (Oak Ridge Isotope Generation and

Depletion) code. This is the recipe used to produce SIMFUEL.

Also listed are WDX (wavelength dispersive x-ray spectrometry)

determined values for the overall and matrix compositions.

These can be compared to determine which components have

formed precipitates and in what amounts. Appendix A. outlines

the procedure for the production of SIMFUEL.







18

The lead-tin-bismuth (33-33-33 wt%) alloy was purchased

in rod form (15"xl/4"xl/4") from Cerro Metal Products of

Bellafonte, Pennsylvania. Table 2.3 lists the grade and

minimum compositions of the individual components used to

produce the alloy. The approximate melting temperature of

this alloy is 2430F (12).


Table 2.3 Grade and minimum composition of the components
used to produce the Pb:Sn:Bi alloy.

Component Grade Minimum Purity
Pb corroding grade 99.94%
Sn grade A 99.80%
Bi 4-9ths 99.99%


The tin-bismuth-gallium (48-48-4 wt%) alloy was produced

by combining the appropriate amounts of the individual

components on a hot plate. The alloy was homogenized for

three hours at 8400F before being cast flat into a chilled

dish. Table 2.4 gives the minimum purity values of the

individual elements used to produce this alloy.


Table 2.4 Grade and minimum composition of the components
used to produce the Sn:Bi:Ga alloy.

Component Minimum Purity
Sn 99.9+%
Bi 99.99%
Ga 99.99%


The tin stock was supplied by Ames Metal Products Co. of

Chicago, Illinois in the form of 1 lb. bars. Pure bismuth was

supplied by Cerro Metal Products in the form of chips. The

gallium was supplied by Eagle-Picher Industries, Inc. of

Quapaw, Oklahoma in a plastic squeeze bottle. The solid







19

gallium was heated in hot water until liquid, then cast into

shot. By doing this, the proper mass required for a batch was

more easily attained. The alumina pellets were supplied by

Coors Ceramic Company and have a OD of 0.320 inches. Tapered

end plugs, made of type 304 stainless steel, were turned on a

lathe to leave a roughened surface for improved mechanical

locking.




2.2 Sample Preparation


Sections of zircaloy-4 cladding approximately six inches

in length were closed at one end with type 304 stainless steel

plugs. Each of the U02 fuel pellets were notched parallel to

the axis with a diamond saw in an attempt to isolate any

liquid metal-UO2 reactions that may have taken place. A

schematic showing the placement of the saw-cut is seen in

Figure 2.1. The width of the notch was 0.012 inches, but the

depth varied from 3/8 to 1/2 the diameter of the pellet.

The specimens were assembled according to the schematic

shown in Figure 2.2. Three different pellets were used in

each specimen. The first pellet, an A1203 segment one inch in

length, provided spacing between the fuel pellet and stainless

steel plug. A third pellet (A1203), approximately two inches

in length, was placed above the U02 pellet as a weight. Solid

pieces of metal alloy were inserted at the bottom and in

between each of the pellets. Force was applied to the top

pellet while heating the sample. As the alloy melted, the












Isolated Region
I


Notch U02 Pellet


Figure 2.1


Schematic showing the location of the notch
placed into UO, pellets prior to testing.















F


o..- ZIRCALOY

SOLID ALLOY

UO 2

AD23

LIQUID METAL


Figure 2.2


Schematic showing the preparation of specimens
for furnace testing.







22

pellets were forced down displacing the liquid metal into the

diametral gap between the pellets and tube.

A stainless steel sleeve containing the prepared specimen

was fitted into one of three sample ports of a furnace shown

in Figure 2.3 and then secured by an o-ring seal. The

schematic in Figure 2.4 shows the position of the specimen

within the furnace apparatus. The samples were evacuated

using a mechanical vacuum pump until a vacuum of approximately

104 torr was reached, after which the samples were lowered

into the hot zone of Barnstead-Thermolyne resistance wound

tube furnaces. Upon reaching temperature, the samples were

backfilled with industrial grade helium (4-5 psi.) to prevent

excessive oxidation of the zircaloy cladding. The samples

were then cycled between vacuum and helium in an attempt to

further reduce the partial pressure of oxygen in the sample

chamber. After a few cycles, a constant positive pressure of

helium was maintained in the stainless steel sleeves during

exposure.




2.3 Testing Times and Temperatures


The times and temperatures of the experiments performed

were chosen in order to correlate the results to those

documented by Adams (8). Experiments were conducted at one of

two temperatures. The first was a low temperature test at

750F used to simulate reactor Standard Operating Conditions

(SOC). These samples were tested for three time periods up to



















































Figure 2.3


Barnstead-Thermolyne tube furnace apparatus.











Vacuum -- -


4m Sleeve




, Specimen


Figure 2.4


Schematic showing the position of the specimen
within the furnace apparatus.


I I









25

5,000 hours. The second was a high temperature test for time

periods of 6 & 24 hours at 15000F, used to simulate a Loss Of

Coolant Accident (LOCA). These temperatures were predicted by

the ESBOND code mentioned earlier.




2.4 Metallographic Preparation


The specimens were prepared and analyzed in cross-

section. Special care was taken to prevent the spread of

radioactive contamination. Preparation involved first

sectioning with a LECO VC-50 diamond saw in cutting oil while

under a dedicated fume hood. Cold mounting with self-setting

resin in one inch molds followed. Polishing was then

performed on a LECO VP-50 polishing wheel under a dedicated

fume hood using the standard metallographic techniques

outlined in Appendix B.




2.5 Characterization Methods


Once polished, the samples were examined using optical

and electron microscopy, electron probe x-ray microanalysis

(microprobe), and Vickers microhardness testing. The optical

microscopy was performed on an Olympus BHM microscope with a

35mm camera and PM-10AD exposure control unit. Optical

imaging was used primarily for making loss of wall thickness

calculations.







26

Electron microscopy was performed on a JEOL 6400 Scanning

Electron Microscope (SEM) with Energy Dispersive Spectrometry

(EDS) ability. The polished samples were imaged in

backscattered mode. Contrast in this mode is created by

differences in the atomic number of the elements present.

When an electron beam is focused on the surface of a solid,

some of the electrons interact elastically with the nuclei of

the atoms. In these elastic collisions, only a small amount

of energy is lost and the electrons are scattered back

(13). The larger the nucleus is, i.e. the higher the

atomic number, the greater the backscattered yield and

brighter the image will appear. EDS provides chemical

information by identifying the characteristic energies of the

x-rays produced during electron-solid interactions (13).

Electron probe x-ray microanalysis (EPMA) was performed

on a JEOL SUPERPROBE 733 using wavelength dispersive

spectrometry. With a lateral resolution of only one micron,

quantitative composition values were obtained for all of the

microstructural features of interest. The instrument compares

the intensity of the sample's emitted x-rays, produced during

electron bombardment, with that of a known standard and

calculates the composition value (13). Both point counts and

line scans were taken on the microprobe. Counts taken in

small microstructural features (close to the lateral

resolution of the instrument) were done individually to ensure

that the information came from within the feature and did not

include information from neighboring areas. Line scans are







27

useful for producing concentration profiles across a region.

This method involves selecting the position of the two end

points followed by the total number of points to be counted.

Line scans were taken across the liquid metal-UO2 interface to

determine if any of the metal components had infiltrated the

UO, pellet.

Vickers microhardness tests were performed in order to

characterize any changes in hardness which may occur to the

components of the fuel over time. A standard of zircaloy-4

containing only solidified alloy was polished and tested.

Hardness values for each component were taken from loads of 10

to 300 grams in order to see if a load dependence exists. No

dependence was observed so a load of 25g. was chosen in order

to produce intentions small enough to measure the hardness of

the solidified alloy within the gap. This load produced large

enough intentions in both the zircaloy and U02 to make

consistent measurements and therefore was used for ease of

testing.




2.6 Average Loss of Wall Thickness Measurements


In an attempt to characterize the dissolution of

zircaloy, loss of tube wall thickness measurements were made

from optical photomicrographs taken at 100x magnification. A

statistical number of measurements were taken from four

photomicrographs of each specimen. The measurements taken

were only of the remaining zircaloy tube and did not include







28

any intermetallic layer that had formed. With each roll of

film, a standard of known length was imaged and used to

calibrate the magnification. By doing this, accurate wall

thickness values for each specimen were calculated and

averaged. Three untested tubes were polished and measured to

provide a true thickness value. The percent loss of tube wall

thickness was calculated using the average (tested) and true

thickness values in the following equation :

((True Tested)/True) x 100 = Percent Loss

The percent loss of tube wall thickness values were

plotted versus testing time to predict the loss over the

lifetime of the fuel. Included on these plots are 95%

confidence intervals which were generated using the two sided

"t" statistic.














CHAPTER 3
RESULTS AND DISCUSSION


3.1 Introduction


The compatibility among liquid metal, UO2, and zircaloy-4

was investigated for two different liquid metal alloys. The

primary alloy studied was Pb:Sn:Bi (33-33-33 wt%), however, an

alloy of Sn:Bi:Ga (48-48-4 wt%) was also the basis of an

abbreviated investigation for improved wetting properties.

Several factors associated with materials compatibility were

evaluated and are discussed below including: intermetallic

compound formation, consumption of zircaloy, and changes in

the properties of the metal alloy and UO over time. In

addition, the compatibility among zircaloy-4, Pb:Sn:Bi, and

SIMFUEL, a simulated high-burnup UO2-based nuclear fuel, was

investigated. These results will be discussed in relation to

service of LBLWR fuel in a reactor environment (section 3.6).




3.2 Intermetallic Compound Formation


From the point of view of qualitative observations of the

microstructure, the formation of intermetallics was rather

striking. Changes in the microstructure of the liquid metal

alloy and at the interface between the liquid alloy and

zircaloy cladding govern the global materials compatibility







30

issues. Therefore, the author has elected to approach this

subject first.

A number of intermetallic compounds have been discovered

in the U02 samples taking primarily two distinct morphologies.

The first is a uniform layer that tends to coat the inside

surface of the zircaloy tubing. The term given to this

morphology will simply be intermetallic layer (IL). The

second morphology is one in which an intermetallic phase

exists in the form of massive particles that are sometimes but

not always faceted (probably idiomorphic in morphology). This

morphology will be termed massive intermetallic (MI). In the

Pb:Sn:Bi samples, time and temperature both play an important

role in determining what phases are present and their relative

abundances.


3.2.1 Pb:Sn:Bi SOC Results


In the low temperature SOC (Standard Operating

Conditions) samples, a number of small MI's with the

composition of ZrSn2 have been observed as seen in Figure 3.1.

In this backscattered electron micrograph, the zircaloy is

shown dark in the bottom right corner. The solidified

structure of the Pb:Sn:Bi alloy is seen upper left. Three

ZrSn2 MI's have formed on the inner surface of the tubing. A

thin IL has started to develop on the surface of the clad.

Although this figure shows far more MI's in one area than is

representative of the fraction of MI's in this sample as a

whole, the number and size of MI's present tends to increase

with increased testing times.
















































Figure 3.1


Backscattered electron micrograph of Pb:Sn:Bi
500 hour SOC sample showing ZrSn2 massive
intermetallic particles.(1,200x)







32

These particles precipitate at the surface of the tubing

as the concentration of Zr in the liquid metal locally exceeds

saturation. Looking at the binary Zr-Sn phase diagram of

Figure 3.2, one sees that the solubility of Zr in Sn is small,

but increases with increasing temperatures (14). These

massive particles were not observed in Adams' bulk samples

consisting of zircaloy tubes filled only with liquid metal

(8). These particles, however, do exist in his samples

containing A1203 pellets. This is a result of the volume of

metal present in the two cases. In the case for the bulk

samples, a large volume of metal is present. This allows for

a larger amount of Zr-dissolution before the concentration

reaches the point where the intermetallics precipitate out.

In the samples containing pellets, the volume of liquid metal

is relatively small, allowing for the precipitation of MI's

from a Zr-saturated alloy in a shorter period of time.

Figure 3.3 shows a backscattered electron micrograph of

a 5,000 hour SOC sample. The UO, is seen light grey at the

top of this micrograph while the zircaloy is shown dark at the

bottom. After this longer time period at temperature, the

MI's have grown larger and have debonded from the surface of

the tubing. The thin ZrSn2 IL developing on the surface of

the clad in the 500 hour sample has thickened in comparison.

A wavelike interface exists in the zircaloy clad where

peaks are observed near the center of these massive

intermetallics. This observation is an indication that the

MI's form at temperature, rather than as a result of


























Atomic Percent T'rin


Zr Weight Percent Tin Sn


Figure 3.2


Zirconium-Tin binary phase diagram. (taken from
ASM Handbook on Alloy Phase Diagrams)




























*' ** 5%- .' *
,

4. .
r

*'v a ** '
**'t ; -.
7. i."-? I

|^^^-low-


.
*"'" -".* :'---uo2
.1 4 .1
.',. "


* zircaloy-4


Figure 3.3


Backscattered electron micrograph of Pb:Sn:Bi
5,000 hour SOC sample.(200x)







35

precipitation due to decreasing solubility concurrent with

cooling. This wavelike interface results from a reduction in

the Zr-dissolution from immediately beneath the MI particles.

Growth of the MI's requires Zr to reach the liquid alloy from

regions between particles; therefore, the zircaloy cladding

appears to be undermined beneath the massive particles

resulting in the wavy pattern. The wavy pattern is under the

control of the Zr-diffusion distance required to reach the

liquid metal. The furthest distances required for Zr to

diffuse around a massive particle are located near the

particles center. Thus, the least amount of dissolution

occurs there and a peak in the zircaloy-4 cladding remains.


3.2.2 Pb:Sn:Bi LOCA Results


A large number of ZrSn2 MI's were observed in the LOCA

samples containing Pb:Sn:Bi as is seen in Figure 3.4. This

figure shows an optical photomicrograph and corresponding

microprobe analysis confirming the composition of the MI's to

be ZrSn2. A thick ZrSn2 IL lines the zircaloy tube (shown at

left). The U02 pellet is seen dark at right. In the figure,

several MI's are observed to align in a belt which runs along

the gap. The variability in Sn composition after 100

micrometers probed occurs as a result of the primary Sn phase

which is located throughout the solidified structure. The

microprobe plot excludes the data for Pb and Bi since neither

of these are involved in the intermetallic formation.

Including these data, the total composition at each point is

raised to 100%.








































T

0 50 100 150 200 250 30
DISTANCE (MICRONS)


-*- Sn -- U -- Zr


Figure 3.4


Optical photomicrograph and electron
microprobe results of Pb:Sn:Bi 24 hour LOCA
sample.(400x)







37

Three different IL's actually exist in these high

temperature LOCA samples and are shown in Figure 3.5. From

top to bottom in this figure are the zircaloy tubing, a thin

Zr4Sn IL, a thin Zr5Sn3 IL, a thick ZrSn2 IL, and the solidified

structure of the Pb:Sn:Bi alloy. Most of the thin Zr5Sn3 IL in

this figure is covered with an oxide formed after polishing.

This oxide is likely the Zr5Sn30o7 phase observed in the related

work of Chapter 1.7. All of these phases are predicted by the

Zr-Sn binary phase diagram as one travels from the zirconium

rich tubing to the "tin rich" alloy. Further discussion of

these intermetallic layers is given in section 3.3.1.

Some of the known physical properties of these

intermetallic compounds are given in Table 3.1 below.


Table 3.1 Material constants of zircaloy-4 and intermetallic
compounds.

Material TLI' Densityb Hardness(hV)c Crystal Structureb
ZrSn2 1142 8.139 237 Orthorhombic
Zr5Sn3 1988 7.420 -- Hexagonal
Zr4Sn 1327 6.078 -- Tetragonal
Zircaloy 1855 6.56 167 Hcp>15900F>Bcc

0 Taken from the binary Zr-Sn phase diagram.(15)
b) Taken from Powder Diffraction Files.
Microhardness values determined with 25 gram load.


Vickers microhardness tests were made to determine the

hardness of the ZrSn2 IL for comparison to that of the

zircaloy tubing. Optical photomicrographs showing

indentations in the ZrSn2 intermetallics and zircaloy clad are

shown in Figure 3.6. In part (a) of this figure, an

indentation is seen within a massive intermetallic particle










































Pb:Sn:Bi
< solidified
.' alloy


Figure 3.5


Backscattered electron micrograph of Pb:Sn:Bi
24 hour LOCA sample showing intermetallic
layers corresponding to phases in Zr-Sn binary
phase diagram.(750x)








I -
)
'B..
U

- A I


Figure 3.6


Optical photomicrograph of Pb:Sn:Bi 24 hour
LOCA sample showing microhardness indentations
in the (A) massive intermetallic particle
(l,000x) and (B) intermetallic layer (500x)


B







40
suspended in the solidified alloy. The value (216 hardness

Vickers) determined for that indention was slightly lower than

the value (237 hV) determined in the IL shown in part (b) of

this figure. The lower hardness value in the MI can be

attributed to two things, cracking of the MI and deformation

of the soft solidified Pb:Sn:Bi alloy during testing.

Three indentations using the same load are seen in part

(b) of Figure 3.6. The lone indentation in the zircaloy-4,

shown bottom right, is larger than the two indentations in the

harder ZrSn2 IL. Since the IL is harder than the zircaloy

tubing, its presence may add to the structural integrity of

the fuel rod. The wear resistance of the clad to abrasion by

the U02 pellet may be improved by the existence of this harder

layer, lowering the risk of a PCI (pellet-cladding

interaction) failure. Also, the hoop strength of the tubes

may be increased slightly. This will depend on the mechanical

properties of the intermetallics at operating temperature as

well as the interfacial strength between these layers and the

zircaloy cladding. In the LOCA tests, samples were cooled

rapidly from 15009F to room temperature producing significant

thermal stresses; however, no cracking or debonding of these

layers was observed in any of the tested samples. This

suggests that the bonding strength between the zircaloy and

intermetallic layers, as well as intermetallics themselves, is

rather good and their presence may benefit the tube integrity

in reactor service.










3.2.3 Sn:Bi:Ga Results


As mentioned earlier, a Sn:Bi:Ga alloy (48-48-4 wt%) also

was studied as a potential bonding material between the UO,

pellets and the zircaloy tubing. Figure 3.7 compares optical

photomicrographs of Sn:Bi:Ga and Pb:Sn:Bi samples tested at

SOC for 500 hours. In these photomicrographs, the UO2 pellets

are shown dark at left. A distinct layer exists in the

Sn:Bi:Ga sample while no apparent layer exists in the Pb:Sn:Bi

sample. Figure 3.8 shows an optical photomicrograph of a 24

hour Sn:Bi:Ga 1,500OF LOCA sample. The zircaloy clad at the

base of the micrograph is jagged as a result of the liquid

metal attack. In both of these figures, a relatively large

portion of the diametral gap is consumed by the formation of

intermetallic compounds consisting of zirconium and either tin

or gallium. The formation of these phases in the amounts

shown will drastically effect the thermal and rheological

properties of the bonding agent and are therefore

unacceptable.

Another shortcoming of this alloy is its apparently high

solubility for UO2, confirmed by the presence of U-Sn-Ga

compounds observed in the saw-cut of a 24 hour LOCA sample.

These compounds are shown in Figure 3.9. Electron beam

microprobe results suggest the composition of the major phase

(shown light grey) is approximately 48wt% U, 48wt% Sn, and

4wt% Ga.
























































Figure 3.7


Optical photomicrographs of (A) Sn:Bi:Ga and
(B) Pb:Sn:Bi 500 hour SOC samples.(100x)


















































Figure 3.8


Optical photomicrograph of Sn:Bi:Ga 24 hour
LOCA sample.(200X)





























" "
Now'


*. X**.1
* ". -. '


Figure 3.9


Backscattered electron micrograph of a
Sn:Bi:Ga 24 hour LOCA sample showing U-Sn-Ga
compounds present at the top of saw-cut:(A)
(700x), (B) (2,000X)










3.3 Consumption of Zircaloy


Thinning of the zircaloy-4 cladding occurs because

zirconium is consumed by the liquid metal at elevated

temperatures. Direct dissolution and alloying/intermetallic

formation are the two mechanisms. A limited solubility of Zr

in the liquid metal exists, although undetectable by Energy

Dispersive Spectrometry (EDS) or microprobe analysis. The

dissolution of Zr into the liquid metal competes with

intermetallic layer formation at the inner surface of the

zircaloy cladding. Each of these mechanisms dominate for a

portion of time during high temperature exposure.


3.3.1 Intermetallic Layer Formation


The ZrSn2 intermetallic layer formed in these samples

does not grow with a planar interface. Instead, the layer

forms a jagged interface with the Pb:Sn:Bi liquid metal alloy

as is displayed in Figure 3.10. This optical photomicrograph

shows a portion of ZrSn2 IL observed at the base of the saw-

cut in a Pb:Sn:Bi 24 hour LOCA sample. In this figure, the IL

is centered between the solidified alloy and the zircaloy

cladding (shown at bottom).

Statistically, perturbations in the intermetallic layer

amplify, generating the spikes observed in this figure. An

inward flux of Zr atoms into the liquid alloy is believed to

reduce the solubility of Sn, forcing it to react with Zr to

form ZrSn.. The ZrSn2 forms between the spikes of the









FUEL ROD CROSS-SECTION


/ FUEL
PELLET
LIQUID
METAL


a
~
,r


Figure 3.10


Optical photomicrograph of Pb:Sn:Bi 24 hour
LOCA sample showing jagged ZrSn2 intermetallic
layer.(1,000x)


CLAD*







47

intermetallic layer where the Zr concentration is the

greatest, causing the layer to thicken and become more uniform

(conversations with Dr. Richard Connell based on his work with

the growth of oxide scales on metal substrates). The Sn may

also react to form ZrSn2 MI's or on the surface of existing MI

particles. The length of time before this phenomenon occurs

depends on the flux of Zr atoms into the Pb:Sn:Bi alloy. The

most important parameters involved are temperature and liquid

metal volume. These directly affect the amount and rate of Zr

solubility in the liquid metal.

The two different stages of intermetallic layer growth

are demonstrated in the optical photomicrograph of a 24 hour

LOCA sample shown in Figure 3.11. This figure displays a

section located near the base of the saw-cut placed in the UO,

pellet. The zircaloy in the left half of this micrograph is

exposed to the larger volume of metal associated with the saw-

cut. As a result, a larger amount of dissolution of Zr can

occur before the solubility of Sn is reduced. Thus, at the

base of the saw-cut, where the liquid metal volume is larger,

a jagged interface exists.

The right half of the micrograph is exposed to a smaller

amount of liquid metal volume, as would be expected in reactor

operation. The concentration of Zr increases more rapidly in

the smaller volume of liquid metal, and as a result, the Sn

has reacted with the Zr to produce a more uniform ZrSn2-liquid

metal interface. As the IL becomes uniform and thickens, both

methods in which the zircaloy is consumed decrease in rate,

















































Figure 3.11


Optical photomicrograph of Pb:Sn:Bi 24 hour
LOCA sample showing change in intermetallic
layer at base of saw-cut.(200x)







49

explaining the flattening off of the loss of tube wall

thickness values over time.

In Figure 3.12, an optical photomicrograph of a 24 hour

Pb:Sn:Bi LOCA sample shows a pronounced amount of tube wall

thinning near the base of the saw-cut; however, a jagged

ZrSn2-liquid metal interface exists. This is consistent with

the proposed theory of intermetallic growth for this system

since the larger volume of liquid metal associated with the

saw-cut requires a greater amount of Zr solution before Sn is

forced to precipitate as ZrSn2.


3.3.2 Loss of Tube Wall Thickness Results


In an attempt to characterize the consumption of zircaloy

by interaction with liquid metal, loss of tube wall thickness

measurements were made from optical photomicrographs taken at

100x magnification. The results for both SOC and LOCA tests

are encouraging. No standard by the Nuclear Regulatory

Commission (NRC) as to the acceptable tube wall thinning under

SOC exists, however, a project goal of less than 5% over the

lifetime of the fuel (40,000 hours) was established. A loss

value of 1.55% was observed in the 5,000 hour SOC sample as is

shown in Figure 3.13. Based on a parabolic rate constant of

1/2, this suggests the project goal will be met.

The loss of wall thickness values for the LOCA samples

are shown in Figure 3.14. Loss values for the 24 hour samples

were less than 6% of the tube wall thickness. The NRC

requires a loss of less than 17% of the tube wall thickness


















































Figure 3.12


Optical photomicrograph of Pb:Sn:Bi 24 hour
LOCA sample showing change in tube wall
thickness at base of saw-cut.(100x)











LOSS OF WALL THICKNESS
S.O.C. SAMPLES


2000 3000
TIME(HOURS)


Figure 3.13


Loss of tube wall
Pb:Sn:Bi SOC samples.


thickness


results of


1000


4000


5000










LOSS OF WALL THICKNESS
L.O.C.A. SAMPLES


6 12 18


TIME(HOURS)


Figure 3.14


Loss of tube wall thickness results of
Pb:Sn:Bi LOCA samples.







53

including external oxidation for LOCA accidents lasting only

one hour (15). The heat stored in the LBLWR fuel is

reduced in comparison to current fuel as a result of improved

heat transfer from the fuel pellets to the cladding via the

high thermal conductivity, liquid metal alloy. The cladding

temperatures reached during a LOCA are significantly reduced

for the LBLWR fuel. Thus, the amount of external oxidation

during a LOCA is expected to decrease over current amounts

more than enough to offset the internal loss to liquid metal

corrosion in the LBLWR design.


3.3.3 Zircaloy-4 Hardness After Exposure


Vickers microhardness tests, approximately 50-80 microns

into the zircaloy, were performed to determine if any changes

in hardness may have resulted from liquid metal attack.

Figure 3.15 shows the results of testing in each of the

components of the Pb:Sn:Bi LBLWR SOC samples. Ninety-five

percent confidence intervals, generated using a two sided "t"

statistic, are included on this graph. A slight decrease in

the hardness of the cladding was observed in these samples.

An even greater drop in the hardness of the cladding was seen

in the LOCA samples as is shown in the graph of Figure 3.16.

This phenomenon is related to an increase in grain size due to

grain growth and was characterized by Adams in his analysis of

anodized specimens (8).










MICROHARDNESS VS TIME
S.O.C. SAMPLES


UO 2
ZIRCALOY










ZIRCALOY


SOLIDIFIED ALLOY


2000


3000


4000


5000


TIME(HOURS)


Figure 3.15


Vickers microhardness results of the Pb:Sn:Bi
LBLWR fuel components in the SOC samples as a
function of testing time.


600.0


500.0


400.0


300.0


200.0


100.0


0.0


1000











MICROHARDNESS VS TIME
L.O.C.A. SAMPLES


... .. .. .. .- -- ---. .. ...-- ------ --- -- ---- --



....ZIRCALOY .





SOLIDIFIED ALLOY
: I t


TIME(HOURS)


Figure 3.16


Vickers microhardness results of the Pb:Sn:Bi
LBLWR fuel components in the LOCA samples as a
function of testing time.


600.0



500.0


400.0



300.0


200.0


100.0



0.0










3.4 Liquid Metal Properties


Any changes that occur to the properties of the liquid

metal over time may radically effect the performance of the

LBLWR fuel and are of considerable interest. Of primary

interest are any changes in the melting point of the bonding

alloy due to variations in composition. Composition

variations could result from dissolution of zircaloy or U02,

as well as loss of Sn during intermetallic formation.

Attempts have been made to perform quantitative metallography

on several of the backscattered electron images in order to

determine the composition of the Pb:Sn:Bi alloy prior to

solidification; however, two of the phases could not be

resolved.

Microhardness tests were conducted on the solidified

alloy to observe any noticeable changes in the hardness which

may ensue from differences in composition. The results of

these tests were reported in Figures 3.15 and 3.16. No abrupt

change in the hardness of the solid alloy was observed.




3.5 UO, Properties


Current data suggests that the UO2 pellet is unaffected

by the liquid Pb:Sn:Bi alloy. Microprobe analysis at the top

of the saw-cut showed no penetration of the alloy components

into the UO2 pellet. Also, no uranium has been discovered in

the Pb:Sn:Bi solidified alloy. As mentioned earlier, uranium







57

was observed in the form of U-Sn-Ga compounds at the top of

the saw-cut in a Sn:Bi:Ga LOCA sample.

The results of microhardness tests conducted in the U02

were reported in Figures 3.15 and 3.16. Microhardness values

were determined 50-80 micrometers within the liquid metal-UO2

interface. No significant change in the hardness of the

ceramic pellet was observed; however, the standard deviation

of the values increased in comparison to the metal components

as a result of the brittle nature of the ceramic oxide.

The formation of an oxide containing zirconium, has been

observed at the surface of the U02 pellet in the high

temperature LOCA samples as well as in the 5,000 hour SOC

sample. This oxide has been observed to coat the entire

surface of the saw-cut in the LOCA samples and is shown black

in Figure 3.17. In the 5,000 hour SOC sample, the oxide phase

was small and scattered along the surface of the U02. The

oxide is seen black while the UO2 appears grey at the top of

the backscattered electron micrograph displayed in Figure

3.18. This oxide phase is sparse in this specimen since at

the lower testing temperature, diffusion and growth rates are

significantly reduced.

The most likely candidate for identification of this

oxide phase is ZrO2, however, it may also include uranium.

Phases containing all of these components were observed in the

related work of section 1.6. The fine scale of these oxide

layers made determining the origin of the EDS signal

difficult. An uncertainty as to whether the signal was



































.4 ..4

^^^'"-'' r'"...
P. PE I- .



."' "" ^ *
._


Figure 3.17


Backscattered electron micrograph of a
Pb:Sn:Bi 24 hour LOCA sample showing zirconium
oxide layer present at the top of saw-cut:(A)
(220x), (B) (700x)


















































Figure 3.18


Backscattered electron micrograph of a
Pb:Sn:Bi 5,000 hour SOC sample showing
zirconium oxide phase present in the saw-
cut. (4,000x)







60

generated only in the oxide layer or included the neighboring

UO2 remains.

Analysis of Adams' samples containing A1203 pellets

revealed no comparable oxide phase present. The enthalpies of

formation of the three oxides--Al203, UO2, and ZrO2--involved in

these two cases provide an explanation. Enthalpy values at

298K as well as approximate values calculated for both SOC

(700K) and LOCA conditions (1,100K) are given in Table 3.2

below (16).


Table 3.2 Enthalpies of Formation of oxides observed in
Pb:Sn:Bi LBLWR fuel samples.(KJ/mole of 0 consumed)

OXIDE Hf (298K)' Hf (700K)b Hf (l,100K)b
ZrO2 -550.7 -538 -526
A1203 -528.0 -515 -499
UO2 -542.6 -404 -244

'' Taken from Thermodynamics in Materials Science (16)
b) Approximate calculation from thermodynamic data.


For all temperatures, the ZrO2 has an enthalpy more

highly negative than the other oxides, suggesting that ZrO2 is

more stable than both UO2 and A1203. Differences in enthalpy

values may be taken as an indication of the driving force to

form the stable species, ZrO2 in this case. As a result of

the greater difference in the case of the U02, a greater

driving force exists and the zirconium reduces the U02, but

not A1203 to form the observed oxide. The difference between

these values is less for the lower temperature SOC samples.

A lower driving force combined with the reduction of diffusion

and growth rates mentioned above, explain the virtual absence

of this oxide in the SOC samples.







61

This reaction likely happens in current reactor

operations once the pellet and tubing become in contact with

one and another. In the LBLWR fuel, the bonding agent allows

this oxide to form prior to pellet-cladding contact, acting as

a conveyor for zirconium.

The existence of this oxide confirms solubility and

diffusion of the Zr from the cladding to the pellet. In the

LOCA samples, the dissolution of the zircaloy was more

pronounced at the surface of the tube near the saw-cut since

a larger volume of liquid metal and a greater surface area of

the pellet exist there. The greater amount of pellet surface

area acts as a sink for dissolved Zr by forming this oxide.

A few of the ZrSn2 MI's have been observed to form in

intimate contact with the UO2 pellets. This only occurred in

the high temperature LOCA samples where the diffusion rates

and solubility limits were high. An example of this is seen

in the backscattered electron micrograph shown in Figure 3.19.

In this figure, the UO2 pellet is located at the bottom of the

backscattered electron micrograph. EDS analysis has revealed

the black phases between the MI and pellet to be zirconium

oxide, which appears to be the precursor to the MI formation

in these rare cases.


















































Figure 3.19


Backscattered electron micrograph of Pb:Sn:Bi
24 hour LOCA sample showing growth of ZrSn2 MI
in intimate contact with UO2.(l,500x)










3.6 LBLWR Compatibility in a Reactor Environment


While this research did not involve studies in a reactor

environment, it is important to consider the effects of

radiation on each of the components in the fuel. By

understanding the system and having knowledge of the changes

that may occur in the presence of radiation, one can make

predictions of changes in materials compatibility that may

result. One well known factor is the production of fission

fragments in the UO2 fuel. SIMFUEL is artificially produced

to simulate the composition of the U02 at the end of life of

the fuel. Thus, studying the compatibility of SIMFUEL in the

LBLWR design, should provide insight as to some of the adverse

reactions among the liquid metal and fission fragments. The

effects of neutron flux are unknown.


3.6.1 SIMFUEL Results


Samples containing SIMFUEL, donated by A.E.C.L. Research

of Ontario, Canada; have been placed into the LBLWR design

containing the Pb:Sn:Bi alloy. SIMFUEL, a simulated high-

burnup UO2-based nuclear fuel, contains rare earth and other

oxides as well as spherical metallic precipitates which exist

in nuclear fuels at the end of life. The six atom percent

burnup samples were investigated for reactions between the

liquid alloy and any of these simulated fission products.

Results suggest no reactions occur between the alloy and these

species. A backscattered electron micrograph of a 1,000 hour







64

SOC SIMFUEL sample is shown in Figure 3.20. The porous

SIMFUEL pellet is shown at the top of this figure.

There does appear to be a difference in the formation of

the zirconium oxide in these samples as opposed to those

studied with natural U02. The SIMFUEL samples show a thicker

and more uniform oxide layer than the U02 samples tested for

an equal time. Initially, this phenomenon was believed to be

attributed to a difference in the surface finish of the two

fuel pellets. The surface of the SIMFUEL pellets appears much

smoother than that of the natural UO2 pellets. Further

examination of the 1,000 hour SIMFUEL sample within the saw-

cut revealed the presence of an equivalent layer which is

shown in Figure 3.21. This denies any dependence of the oxide

formation on surface finish.


3.6.2 Compatibility Changes Under Radiation


The effects of radiation on the intermetallic formation

and growth in the Pb:Sn:Bi LBLWR fuel is difficult to

speculate. There is no reason to believe that any changes in

these processes will occur in a reactor environment.

Radiation is known to break up protective coatings on metallic

materials, increasing the corrosion of these materials by

subjecting freshly exposed metal to further attack. Due to

the apparently strong bonds between the zircaloy-4 and

intermetallics in these samples, however, it would not be

expected to occur.




























F; ': ?* *? ** *
- -* -*>: *
^-s:^- -.-*,? -. '* -'-0 ** *
^ .'* ** '-^ -* "* "-


Figure 3.20


Backscattered electron micrograph of Pb:Sn:Bi
1,000 hour SOC SIMFUEL sample.(360x)















































g a '0"
,, a 0 --. *
@- m .U .9
Sge -, 0
9 S. S-* **~ *


49

'It:
Il


i.r


*

* *
,, "
*"


0
e

S.


Backscattered electron micrograph of Pb:Sn:Bi
1,000 hour SOC SIMFUEL sample showing
zirconium oxide within the saw-cut.(1,000x)


U


Figure 3.21


Vo















CHAPTER 4
CONCLUSIONS


4.1 Introduction


The objectives of this research were: 1) to characterize

the materials compatibility of the liquid metal bonded LWR

fuels using optical and electron microscopy, electron probe x-

ray microanalysis, and Vickers microhardness testing, 2) to

predict the compatibility of the liquid metal bonded fuels in

an actual reactor environment with the aid of studies

involving SIMFUEL, and 3) to suggest topics for future

investigation. The following conclusions are drawn from this

research.




4.2 Results of Materials Compatibility for Liquid Metal
Bonded Light Water Reactor Fuel Elements


Thinning of the zircaloy-4 cladding occurs because

zirconium is consumed by the liquid metal at elevated

temperatures. Direct dissolution and alloying/intermetallic

formation are the two mechanisms.


4.2.1 Loss of Tube Wall Thickness Results


1) An average loss of tube wall thickness value of 1.55%

was observed in the 5,000 hour sample under SOC conditions

when using the Pb:Sn:Bi alloy (33-33-33 wt%).







68

2) Average loss values for the 24 hour Pb:Sn:Bi samples

under LOCA conditions were less than 6% of the tube wall

thickness.

3) These results suggest that the Pb:Sn:Bi alloy

continues to meet the compatibility requirements of this

application when in the presence of U02.


4.2.2 Intermetallic Compound Formation


1) Intermetallic compounds having the morphology of

layers (IL) and massive intermetallic particles (MI) form in

the LBLWR samples containing both Pb:Sn:Bi and Sn:Bi:Ga (48-

48-4 wt%) liquid metal alloys.

2) Massive intermetallics having a composition which

corresponds to ZrSn2, precipitate from a Zr-saturated Pb:Sn:Bi

alloy under both testing conditions.

3) Three intermetallic layers develop on the surface of

the zircaloy-4 cladding. These layers correspond to the

following compounds: Zr4Sn, ZrsSn3, and ZrSn2. The interfacial

strength between these layers and the zircaloy-4 clad appears

rather good.

4) A wavy zircaloy-ZrSn2 interface develops in the SOC

samples containing Pb:Sn:Bi due to a reduced Zr dissolution

beneath the barrier MI particles.

5) A jagged ZrSn2-Pb:Sn:Bi liquid metal interface

develops by statistical amplification of surface

perturbations. As the Zr concentration increases in the

Pb:Sn:Bi alloy, the Sn solubility is reduced, forcing Sn to







69

react to form ZrSn2. This reaction takes place between the

spikes of the intermetallic layer where the Zr-concentration

is greatest, causing the layer to thicken and become more

uniform.

6) The Sn:Bi:Ga alloy has a relatively high solubility

for uranium, confirmed by the formation of U-Sn-Ga compounds

in the LOCA samples containing this alloy.

7) A relatively large portion of the diametral gap in

the Sn:Bi:Ga samples was consumed by the formation of

intermetallic compounds consisting of zirconium and either tin

or gallium. The combination of these two phenomenon--uranium

solubility and excessive intermetallic formation--render the

Sn:Bi:Ga alloy unacceptable for this application.


4.2.3 Results of Hardness Testing


1) No change in the hardness or microstructure of the

solidified Pb:Sn:Bi alloy suggest that it remains liquid at

operating temperature with no major variations in composition.

2) A significant reduction in the hardness of the

zircaloy-4 cladding tested under LOCA conditions is a result

of recrystalization and grain growth within the

microstructure.

3) The ZrSn2 intermetallic layer is harder than the

zircaloy-4 cladding at room temperature and may add to the

tube integrity in reactor service.










4.2.4 UO2 Compatibility in the Pb:Sn:Bi LBLWR Fuel


1) The Pb:Sn:Bi liquid metal alloy has no solubility for

UO2 at these testing temperatures, and does not embrittle the

fuel material by penetration into the UO2 microstructure.

2) The Pb:Sn:Bi alloy acts as a medium for Zr transport

to the surface of the pellet, where it reduces the U02 to form

a zirconium rich oxide. The most likely candidate for

identification of this oxide phase is ZrO2, however, it may

also include uranium.



4.3 Prediction of LBLWR Compatibility in
a Reactor Environment

1) No major changes or reactions were observed in the

Pb:Sn:Bi LBLWR samples containing SIMFUEL.

2) A thicker and more uniform zirconium oxide layer

develops in the SIMFUEL samples tested under SOC conditions

for the same time periods, however, this is not a result of

differences in surface finish.

3) There is no reason to believe that the processes

occurring in these samples will change under the influence of

radiation.



4.4 Topics for Future Investigations


1) The current tests containing Pb:Sn:Bi with both

SIMFUEL and A1203 pellets should be carried out to at least









25,000 hours.

2) The design of a LBLWR fuel rod is required before

investigations in a test reactor can be conducted. This

design must account for the reduction in fission gas space as

a result of the incompressible liquid metal alloy present.

3) LBLWR fuel exposure within a test reactor is the last

major step in confirming the compatibility of Pb:Sn:Bi as the

thermal bonding material of this fuel design. The

interactions between the UO2 pellets and intermetallic phases

as the radial gap closes will be of great importance.

4) Other areas for investigation which would provide

useful information include:

a) determination of the ZrSn, mechanical properties

at elevated temperatures.

b) determination of the interfacial strengths

between the zircaloy-4 and intermetallic compounds.















APPENDIX A
PREPARATION OF SIMFUEL



1. Vacuum-dried, high-purity additives (99.999% pure

oxides listed in Table 2.2) were dry mixed with fine urania

powder.

2. High-energy, wet, stirred-ball milling was used to

produce a uniform submicrometer dispersion.

3. An alcohol based slurry was stirred magnetically to

keep a suspension.

4. The suspended slurry was spray dried locking the

selected composition into granules.

5. Conventional methods were used to compact and press

the granules into a green ceramic pellet.

6. Sintering at 16500C for 2 hours under flowing H2

allows the system to reach equilibrium and dispersion or

precipitation to occur.















APPENDIX B
METALLOGRAPHIC PREPARATION



The mounted samples were polished using the following

procedure:

1. On a LECO VP-50 circular polishing wheel operating at

250 rpm., the samples were hand ground with 180 grit Sic paper

until the entire surface was ground flat. The sample was held

in one spot to create one pattern of grooves. Note: due to

the hazardous material, plastic gloves were worn.

2. The samples were washed using liquid dish detergent

while rubbing lightly with a finger to remove any SiC which

may have attached itself to the surface.

3. Steps one and two were repeated for 240, 320, 400,

600, 800, and 1200 grit Sic papers. The samples were held

such that the new grooves created were rotated 900 from those

created by the previous paper. Each polishing step lasted

approximately 45-60 seconds.

4. After the grinding stages, diamond polishing was

performed on pan "w" using water soluble lapping oil. A rough

polish using 6 micron diamond for 3-5 minutes was completed

first. After washing, a fine polish using 1 micron diamond

followed.

5. After the final wash, the sample was rinsed with

reagent alcohol and blown dry.















LIST OF REFERENCES


1. R.F. Wright, Liquid Bonded Light Water Reactor Fuel:
Enhanced Light Water Reactor Safety and Performance,
Doctoral Dissertation, University of Florida,
Gainesville, FL (1994).

2. ESCORE--The EPRI Steady-State Core Reload Evaluator Code:
General Description, EPRI NP-5100, prepared by Combustion
Engineering, Inc., Palo Alto, CA (February 1987)

3. J. S. Tulenko, "An Innovative Fuel Design Concept for
Improved Light Water Reactor Performance and Safety",
unpublished paper, University of Florida, Gainesville, FL
(1991).

4. J.P. Ligou, Elements of Nuclear Engineering, (Harwood
Academic Publishers, New York, NY, 1986) p.210.

5. J.S. Tulenko, R.F. Wright, G.J. Schoessow, T. Adams, R.G.
Connell, "Material Performance Evaluation of the
Innovative Liquid Metal Bonded Light Water Reactor Fuel
Rod", Proceedings of the International Topical Meeting on
LWR Fuel Performance, (American Nuclear Society, West
Palm Beach, FL, April 17-21, 1994).

6. J.R. Davis, J.D. Destefani, and H.J. Frissell, Editors,
Metals Handbook Ninth Edition Volume 13: Corrosion, (ASM
International, Metals Park, OH, 1987), p.92-96.

7. E.C. Miller, "Corrosion of Materials by Liquid Metals",
Liquid Metals Handbook, (Washington D.C., 1952), p.144-
145.

8. T. Adams, Feasibility Study of Materials Compatibility
For Liquid Metal Bonded Light Water Reactor Fuel
Elements, Master's Thesis, University of Florida,
Gainesville, FL (1994).

9. W. Rostoker, J.M. McCaughey, H. Markus, Embrittlement by
Liquid Metals, (Reinhold Publishing Corporation, New
York, NY 1960), p.19-32.

10. C.O. Smith, Nuclear Reactor Materials, (Addison-Wesley
Publishing Company, Reading, MA 1967) p.59-86









11. P.G. Lucuta, R.A. Verrall, Hj. Matzke, B.J. Palmer,
"Microstructural Features of SIMFUEL Simulated High-
Burnup UO2-Based Nuclear Fuel", Journal of Nuclear
Materials, 178, (1991), p.48-60

12. J. R. Weeks, D. H. Gurinsky, "Solid Metal-Liquid Metal
Reactions in Bismuth and Sodium," Proceedings of the
Thirty-Ninth National Metal Congress and Exposition
(American Society of Metals, Chicago, 1957).

13. K. Mills, J.R. Davis, J.D. Destefani, D.A. Dieterich,
G.M. Crankovic, and H.J. Frissell, Editors, ASM Handbook
Volume 10: Materials Characterization, (ASM
International, Metals Park, OH, 1986)

14. H. Baker and H. Okamoto, Editors, ASM Handbook. Vol. 3:
Alloy Phase Diagrams, (ASM International, Metals Park,
Ohio, 1992) p.2*372.

15. Code of Federal Regulations, "10 CFR 50.46", (U.S,
Government Printing Office, Washington D.C. 1993)

16. R.T. DeHoff, Thermodynamics in Materials Science,
(McGraw-Hill, Inc., New York, NY, 1993)















BIOGRAPHICAL SKETCH


The author was born on April 9, 1970, in Willoughby,

Ohio. He received his elementary education in public schools

located in Mentor, Ohio, and St. Cloud, Florida. He graduated

from St. Cloud High School in June of 1988 and immediately

began attending the University of Florida in July. While at

the University of Florida, the author pursued a Bachelor of

Science in materials science and engineering, eventually

graduating with honors in May of 1993. In August of 1993, the

author enrolled in the graduate materials science and

engineering program at the University of Florida. While in

graduate school, the author worked as a graduate assistant to

Dr. Richard G. Connell, Jr. In May of 1995, the author plans

to receive his Master of Science degree and begin his career.

The author is married to the former Miss Samantha Kelly

Lyons of St. Cloud, Florida. He enjoys watching and playing

several different sports including basketball, golf, and

football.









I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a thesis for the degree of Master of Science.


Richard G! Connel Jr.,
Committee Chairman
Associate Professor of Materials
Science and Engineering

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a thesis for the degree of Master of Science.


Robert T. DeHaC
Professor of Materials Science
and Engineering

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a thesis for the degree of Master-of Science.


Paul H. Holloway
Professor of Materials Science
and Engineering

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope n quality, as
a thesis for the degree of Mas of Science


Ji s S. Tulenko
Pressor of Nuclear Engineering
Sciences

This thesis was submitted to the Graduate Faculty of the
College of Engineering and to the Graduate School and was
accepted as partial fulfillment of the requirements for the
degree of Master of Science.

May, 1995 /, ----_ --
Winfred M. Phillips
Dean, College of Engineering



Karen A. Holbrook
Dean, Graduate School
















1I80
199j













UNIVERSITY OF FLORIDA
111111 I II I I III511 t111111 IIII
3 1262 08554 5514




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EWK9L1LXZ_HYU6R1 INGEST_TIME 2012-02-20T21:41:44Z PACKAGE AA00009021_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES