Studies of laser-target interactions in pulsed excimer laser evaporation of superconducting oxides and other metal oxides

MISSING IMAGE

Material Information

Title:
Studies of laser-target interactions in pulsed excimer laser evaporation of superconducting oxides and other metal oxides
Physical Description:
xxiii, 301 leaves : ill. ; 29 cm.
Language:
English
Creator:
Thé, Stephens Sin-Tsun
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1993.
Bibliography:
Includes bibliographical references (leaves 286-300).
Statement of Responsibility:
by Stephens Sin-Tsun Thé.
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 - 001938260
notis - AKB4396
oclc - 30947320
System ID:
AA00003681:00001

Full Text







STUDIES OF LASER-TARGET INTERACTIONS IN PULSED EXCIMER LASER
EVAPORATION OF SUPERCONDUCTING OXIDES AND OTHER METAL OXIDES















By
STEPHENS SIN-TSUN TE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1993


UNIVERSITY OF FLORIDA LIBRARIES























Copyright 1993

by
Stephens Sin-Tsun Th6
































To My Wife,
Jennifer










ACKNOWLEDGEMENTS


I would like to express my gratitude to Professor Tim Anderson, my thesis advisor,
for his insight and effort have made this research possible. His numerous trips to
Wilmington, DE, during the progress of this research were greatly appreciated.

I am grateful to Dr. Ken Keating, my mentor and thesis coadvisor. His patience
and guidance during my internship at the Experimental Station were greatly appreciated.
His insight and wisdom as as a research fellow at DuPont are invaluable in my industrial
training. Also his efforts in maintaining funding supports during difficult times are greatly
appreciated.

E.I. du Pont de Nemours and Co., especially Drs. Ken McKelvey, Bob Dorothy,
Lew Goodrich, and Jim Trainham, are thanked for providing me with the research funding
and the opportunity to work in this great research environment, while allowing me to learn
the intricacy of industrial research.

The excimer laser group, Dr. Jim Hohman, Sandy Witman, Gerry Frost, and Don
Wonchoba, are thanked for putting up with me for all these years and for kindly providing
me with a lot of help and inspiration. This research would not have been possible without
their expertise in optics and lasers.

The superconductivity group, especially Drs. Bill Holstein, Dean Face and Dan
Laubacher, are thanked for providing me with constructive discussions and assistance in
sample analysis.

John Pedrick is thanked for his SEM expertise and for providing me with all the
SEM and EDAX analysis in this research; without his help this dissertation would have
been deprived of pictures. Al Terzian is thanked for helping me with the design and
construction of the laser deposition chamber. Dr. Raj Bordia, Rajesh Patel, and Phil Dye
are thanked for the preparation of ceramic samples. Larry Harrison is thanked for the
thermal analysis. Bert Diemer is thanked for providing some thermodynamic data. Larry
Eichelberger is thanked for his photographical expertise. Dr. Rudy Enck at British
Petroleum Research is thanked for the work on laser thermal diffusivity measurement Carl
Mueller from the University of Florida is thanked for the Auger analysis.







Researchers and staff initially in the Engineering Technology Laboratory, then in
Central Research and Development, Science and Engineering, and lately in Corporate
Science and Engineering, including Drs. Kurt Mikeska, Deepak Doraiswamy, Juan
Figueroa, Mike Kelley, and many more, are thanked for providing me with valuable
discussions and support.

Staff and colleagues at the Chemical Engineering Department, University of
Florida, especially Shirley Kelly and Roger Aparicio, are thanked for helping me with
paperwork and registration while I was off-campus.

My parents and family are thanked for their endless supports and encouragements;
and for their financial support while I was working on my undergraduate degree. Also I
am thankful to the Lord for His benediction.

I would like to give special thanks for my loving wife, Jennifer, for her infinite
patience and understanding while I was completing my graduate study and for her
assistance in the preparation of this manuscript.











TABLE OF CONTENTS



ACKNOWLEDGEMENTS ...................................................................... iv

LIST OF TABLES ............................................................... ........... xiii

LIST OF FIGURES ..................................................... .................... xiv

ABSTRACTS................................................................................. xxii

CHAPTERS

1. INTRODUCTION.................................... ........................... 1

1.1 Overview of High Tc Superconductor Thin Film............................. 3
1.1.1 Status and Applications.................................. ......... 3
1.1.1.1 Superconductor and its properties....................... 3
1.1.1.2 Passive microwave devices............................... 7
1.1.1.3 Active microwave devices................................9
1.1.1.4 Integrated Systems........................... .......... 11
1.1.2 HTSC Thin Films .................................................. ... 13
1.1.2.1 Ideal film properties...................................... 13
1.1.2.2 Characterization techniques ............................... 13
1.1.2.3 Substrates .................................................. 15
1.1.2.4 Technological challenges.................................. 16
1.1.3 Comparison of Techniques for Deposition of YBCO............. 16
1.1.3.1 Electron-beam evaporation................................ 16
1.1.3.2 Sputtering .................... ......................... 17
1.1.3.3 Metallorganic chemical vapor deposition (MOCVD) .... 17
1.1.3.4 Pulsed-laser deposition (PLD)........................... 18
1.2 Overview of Pulsed-Laser Evaporation......................... ............. 18
1.2.1 Hardware Configurations ........................... ............ .. 18
1.2.2 Operational Procedures ............... ........................... 20
1.2.3 Typical Results ....................................................... 22






1.2.4 Overview of Laser-Target Interactions.............................. 23
1.2.4.1 Photon absorption and phase transformation ......... 24
1.2.4.2 Plasma expansion ...................................... 25
1.2.5 Some Outstanding Problems for PLD Technique .............. 26
1.2.5.1 Boulder generation....................................... 26
1.2.5.2 Target stoichiometry and texturing..................... 26
1.2.5.3 Process scale-up............................. ........... 27
1.2.5.4 Substrate-film interface reactions....................... 27
1.3 Overview of This Work........................................................... 28
1.3.1 Target Surface Morphology........................................ 28
1.3.2 Particulate Reduction.............................. ............. 28
1.3.3 Target and Film Stoichiometry.................................29

2. EQUILIBRIUM ANALYSIS OF THE YBCO SYSTEM .........................30

2.1 Introduction...................................... ............................. 30
2.1.1 Chemical Equilibrium Analysis...................................... 32
2.1.2 Fundamental Assumptions ........................................... 33
2.1.2.1 System temperature .................................... 33
2.1.2.2 Local equilibria ............................................ 34
2.1.2.3 Charge neutrality .......................................... 35
2.1.2.4 Electronic excitation..................................... 36
2.1.3 Theoretical Backgrounds............................................... 37
2.1.4 Outline of the Computer Program..................................... 38
2.1.5 Statement of Purpose ............................................... 40
2.2 Thermodynamic Calculations................................. ........... 40
2.2.1 Data Fitting Functions ............................................. 40
2.2.2 Condensed Phases .............................. ................... 41
2.2.3 Gas Phase ............................................................. 47
2.2.4 Oxygen System.................................................. 47
2.2.5 Yttrium-Oxygen System............................................. 50
2.2.6 Barium-Oxygen System.................................. ........... 52
2.2.7 Copper-Oxygen System................................ ........... 52
2.3 Complex Reaction Equilibria at Elevated Temperature.........................55
2.3.1 Effect of Temperature...............................................55
2.3.2 Effect of Pressure .............................................. 56






2.3.3 Effect of Oxygen Pressure ............................................ 59
2.3.4 Effect of Inert Gas.......................................... ....... 68
2.3.5 Analysis of PLD Plasma.................................. ...... 68
2.4 Conclusion ........................................................................70

3. SELECTED THERMOCHEMICAL AND THERMOPHYSICAL
PROPERTIES STUDIES..............................................................71

3.1 Introduction ....................................................................... 71
3.2 Sample Preparation................................................................72
3.3 Differential Thermal Analysis ............................................... 73
3.3.1 Introduction............................................................ 73
3.3.2 Experimental........................................................... 73
3.3.3 Result and Discussion .................................................74
3.4 Thermal Gravimetric Analysis................................. ............. 78
3.4.1 Introduction.............................................................78
3.4.2 Experimental........................................................... 78
3.4.3 Result and Discussion ................................................79
3.5 Differential Scanning Calorimetry ............................................ 79
3.5.1 Introduction..............................................................79
3.5.2 Experimental............................................................ 81
3.5.3 Result and Discussion ............................................... 81
3.6 Laser Flash Thermal Diffusivity Measurement.................................. 81
3.6.1 Introduction............................................................. 81
3.6.2 Experimental............................................................ 84
3.6.3 Result and Discussion ................................................ 84
3.7 Thermal Mechanical Analyzer....................................................85
3.7.1 Introduction............................................................. 85
3.7.2 Experimental............................................................. 90
3.7.3 Result and Discussion ................................................90
3.8 Conclusion .................................................................. ...90

4. THERMAL MODELS.................................................................. 92

4.1 Introduction ......................................................................92
4.2 Important Phenomena and Process Description............................... 94
4.2.1 Surface Morphology ..................................................95







4.2.2 Thermal Penetration Depth ............ ............................. 95
4.2.3 Plasma Kinetic Energy................................. ........... 96
4.2.4 Physical Definition of Process to be Modeled ..................... 97
4.3 Review of Laser Heating Model .................... .... ......... .......... .. 98
4.3.1 Theory of Electromagnetic Radiation ............................... 98
4.3.1.1 Electromagnetic radiation................................... 98
4.3.1.2 Surface and absorption...... ........................... 99
4.3.2 Laser Activated Heat Transfer Models ......... ........ ........... 101
4.3.2.1 Fundamental models....................................... 101
4.3.2.2 Surface temperature................................... ...... 104
4.3.2.3 Subsurface heating......................................... 105
4.3.2.4 Heating and cooling rates .................................. 106
4.3.2.5 Heat transfer with material ejection ...................... 106
4.4 Macroscopic Energy Balance................................................... 107
4.4.1 Photon Absorption................................................. 108
4.4.2 Phase Transformation............................ ................... 108
4.4.3 Radiation, Convection and Thermal Conduction Losses ........ 110
4.4.4 Plasma Kinetic Energy................................................ 112
4.5 Conclusion ......................................................................... 112

5. PARTICLE PRODUCTION STUDIES .............................................. 114

5.1 Overview of Particulate Formation ............................................. 114
5.1.1 Introduction........................................................... 114
5.1.2 Surface Roughness..................................................... 115
5.1.3 Formation of Clusters................................................. 116
5.1.4 Formation of Particulates............................................ 117
5.2 Effect of Deposition Parameter Film Deposition............................. 119
5.2.1 Effect of Laser Parameter....................................... 119
5.2.2 Effect of Deposition Parameter ...................................... 121
5.2.3 Effect of Target Quality............................................... 122
5.3 Techniques of Particulate Reduction..................................... 122
5.3.1 Secondary Laser..................................................... 122
5.3.2 Mechanical Velocity Filter........................................... 123
5.3.3 Surface Heating .......... ... ............................................ 123
5.3.4 High-Velocity Target and Plasma Spray........................... 124






,5.3.5 Ag-Doped Target .................................................. ..... 124
5.3.6 Off-Axis Deposition................................................. 124
5.3.7 Substrate Negative Voltage Bias..................................... 124
5.4 Particulate Reduction via Target Heating....................................... 125
5.4.1 Introduction............................................................ 125
5.4.2 Experimental Procedure ............................................... 126
5.4.3 Results and Discussion.......................................... ....... 128
5.4.3.1 Droplet temperature and liquid molten pool............. 136
5.4.3.2 Surface morphology and droplet formations ............. 144
5.4.3.3 Hydrodynamic sputtering and particle size............... 145
5.5 Conclusion ......................................................................... 147

6. TARGET MORPHOLOGY OF YBCO SUPERCONDUCTOR...................... 148

6.1 Review of Literature............................... ....................... 148
6.1.1 Introduction......................................................... ..... 148
6.1.2 Surface Modification.................................................... 149
6.1.3 Target Morphology in YBCO......................................... 149
6.1.4 Statement of Purpose .................................................. 150
6.2 Target Morphology Studies of YBCO Systems............................... 150
6.2.1 Experimental Setup.................................................... 151
6.2.2 Effects of Laser Fluence............................................ 154
6.2.2.1 Porous....................................................... 159
6.2.2.2 Cones........................................................ 169
6.2.2.3 Melts.......................................................... 176
6.2.3 Effect of Pulse Number..................................... .... 183
6.2.3.1 Melting of surface: less than 20 pulses................... 183
6.2.3.2 Formation of columns: 20 to 100 pulses................. 186
6.2.3.3 Formation of tables: higher than 100 pulses.......... 186
6.2.4 Effect of Phase Impurities............................................ 192
6.2.5 Effects of Repetition Rate and Beam Size.......................... 201
6.2.6 Effects of Environments................ .............................. 201
6.2.7 Effects of Target Density ......................................... 207
6.3 Target Morphology Studies of Other Materials................................ 212
6.3.1 Ceramics............................................................... 212
6.3.1.1 LaA103 ..................................................... 212







6.3.1.2 A203 ......................................................... 214
6.3.2 M etals ......... ...... ............................................... ... 214
6.3.2.1 Titanium..................................................... 214
6.3.2.2 Gold and copper.......................................... 218
6.3.3 Polymers .............................................................. 222
6.4 Modelling on Target Morphology ........................ ...... ..........222
6.4.1 Surface Morphology at Low and Medium Fluences ............... 224
6.4.1.1 Radiation heating and surface asperties .. ..... ............ 227
6.4.1.2 Shielding of porous and cones............................ 228
6.4.2 Surface Morphology at High Fluence................................ 229
6.4.2.1 Surface cracks due to self-stress ........................ 231
6.4.2.2 Surface tension induced ripples ...........................235
6.4.2.3 Fluence-dependent optical analysis ...................... 236
6.5 Conclusions ........................................................................ 239

7. PLUME EXPANSION STUDIES ..................................................244

7.1. Overview of Laser Induced Shock Wave................................... 244
7.1.1 Photon Absorption and Induction Period........................... 245
7.1.2 Fundamental of Shock Wave Formation ............................245
7.1.3 Laser-Induced Shock Wave........................................ 247
7.1.4 Recoil Pressure and Plume Redeposition........................... 249
7.2 Plume Expansion Studies of YBCO Systems............................... 249
7.2.1 Statement of Problem................................................. 249
7.2.2 Experimental Details ... ...... ...... ....... .......................... 250
7.2.3 Effect of Fluence ................................................. 250
7.2.4 Effect of Geometry..................................................... 250
7.2.5 Effect of Ambient Pressure........................................... 258
7.3 Conclusions ....................................................................... 258

8. MASS SPECTROSCOPIC INVESTIGATION..................................... 259

8.1 Review of Literature................................................................ 259
8.1.1 Plasma Characterization .............................................. 259
8.1.2 Mass Spectroscopy Fundamentals...................................261
8.1.3 Mass Spectroscopic Plasma Analysis................................. 262
8.1.3.1 Atoms and molecules..................................... 263





8.1.3.2 Clusters..................................................... 263
8.1.3.3 Quantitative determination................................. 264
8.1.3.4 Kinetic energy ................................................ 264
8.1.3.5 Photochemistry........................................... 264
8.2 Mass Spectroscopic Analysis of YBCO ...................................... 265
8.2.1 Experimental........................................................... 265
8.2.2 Effect of Fluence ...................................................... 268
8.2.3 Effect of Ambient Gas and Pressure....................................... 272
8.2.4 Effect of Repetition Rate ............................................... 276
8.2.5 Effect of Pulse Number ............................................ 278
8.2.6 Effect of Target Density .................................................. 278
8.3 Comparison with Complex Equilibrium Reactions...........................278
8.3.1 Effect of Temperature............................................. 279
8.3.2 Effect of Pressure ................................................ .. 279
8.3.3 Effect of Background Gas............................................ 280
8.4 Conclusion ............................................................... ..... 280

9. CONCLUSION............................................................... ..... 281

9.1 Reduction of Film Particulate ................................................... 281
9.2 Development of Target Morphology ..............................................282
9.3 Chemical Equilibrium Analysis.................................................. 283
9.4 Recommendation for Future Studies............................................ 284

REFERENCE LIST.............................................................................. 286

BIOGRAPHICAL SKETCH .................................................................. 301











LIST OF TABLES


Table page
2-1 Thermodynamic coefficients of the species in YBCO chemical
equilibria 43
2-2 Solid phases of YBCO system 44

2-3 List of vapor species detected in laser induced YBCO plasma 48

2-4 Pressure dependence on the boiling point temperature of copper
element 58

2-5 Equilibrium composition of YBCO vapor phase at 8000 K and 1 atm
with various initial conditions 67

3-1 Temperature rise time and thermal diffusivity of high density pellets
(94 % theoretical density) irradiated using a C02 laser. Temperature
was measured at the rear surface. 87

3-2 Temperature rise time and thermal diffusivity of low-density pellets
(65% theoretical density) irradiated using a CO2 laser. Temperature
was measured at the rear surface. 88

5-1 Comparison of the number of particulates generated by laser
deposition taken from this work and literature 131

6-1 Morphology of YBCO target evaporated using a KrF laser at fluence
of 0.1 to 2.0 J.cm2 for 2000 pulses in 1 atm air. Evaporation size
of 100 l m x 100 lp m 156

6-2 Surface morphological development of high- and low-density
YBCO target evaporated using a KrF laser in 100 mTorr oxygen._ 208

8-1 Atomic species detected using mass spectroscopy at fluence of 0.3 -
3.0 J/cm2. 271

8-2 Effect of pressure on the composition of YBCO plasma species 275

8-3 Effect of fluence on the composition of YBCO plasma species 277











LIST OF FIGURES


Figure P=g

1-1 YBa2Cu307 superconductor (a) atomic structure; (b) temperature
dependent resistivity. 6
1-2 Schematic view a stripline resonator. (a) cross section showing
conductor and dielectric; (b) top view. 8

1-3 Schematic view a HTSC Josephson Junction circuit. 10

1-4 Cross sectional view a integrated SQUID magnetometer (a) on
LaA103 substrate; (b) on YSZ substrate. 12

1-5 Schematic diagram a pulsed laser deposition chamber. Bottom view
shows zones of laser-target interactions. 19

2-1 Temperature dependent heat capacity of O+. Comparison of three
different fitting functions. 42

2-2 Gibbs free energy yttrium element as a function of temperature. 46

2-3 Oxygen phase equilibria at 1 atm. 49

2-4 Yttrium-oxygen vapor phase chemical equilibria at 1 atm. 51

2-5 Barium-oxygen vapor phase chemical equilibria at 1 atm. 53

2-6 Copper-oxygen vapor phase chemical equilbria at 1 atm. 54

2-7 YBCO vapor phase equilibria at temperatures from 2500 to 20000
K and pressure 1 atm. 57

2-8 YBCO vapor phase equilibria at temperatures from 5000 to 10000
OK and pressure 0.2 atm. 60

2-9 YBCO vapor phase equilibria at temperatures from 5000 to 10000
K and pressure 0.5 atm 61

2-10 YBCO vapor phase equilibria at temperatures from 5000 to 10000
oK and pressure 1 atm 62

2-11 YBCO vapor phase equilibria at temperatures from 5000 to 10000
OK and pressure 10 atm. 63







2-12 YBCO vapor phase equilibria at temperatures from 5000 to 10000
oK and pressure 50 atm. 64

2-13 Mole fraction and number density of electron in YBCO plasma at a
temperature of 8000 K and pressures up to 50 atm. 65

2-14 Electron density of YBCO plasma at temperatures of 5000, 8000
and 10000 K at pressures up to 50 atm. 66

2-15 Composition of YBCO plasma at temperatures between 5000 and
10000 K and pressures up to 50 atm. Note the formation of oxides
at low temperature and high pressure and the formation of ions at
low pressure and high temperature. 69

3-1 Differential Thermal Analysis of YBCO powder. Heating rate: 20
oC/min. Sample size: 39.2 mg. Temperature range: 30 to 1600 OC.
Atmosphere: air. 75

3-2 Differential Thermal Analysis YBCO pellet (wt. 99.9 mg), YBCO
powder (wt. 39.2 mg) and gold powder (wt. 86.9 mg) at
temperatures around the melting point. Heating rate: 20 oC/min.
Atmosphere: air. 76

3-3 Resolidified sample of YBCO powder after undergoing DTA
analysis to 1100 C. Magnification (a) 30x; (b) 600x 77

3-4 Thermal Gravimetric Analysis of YBCO powder. Heating rate: 10
oC/min. Sample size: 44.9 mg. Atmosphere: N2. 80

3-5 Differential Scanning Calorimetry YBCO. Heating rate: 10 oC/min.
Temperature range: 30 to 600 oC. Atmosphere: N2. Peak at 270 oC
is probably due to organic impurity. 82

3-6 Temperature rise during laser flash technique used in determining
thermal diffusivity. 86

3-7 Thermal diffusivity YBCO as a function temperature, measured up
to 400 oC. 89

3-8 Thermal Mechanical Analysis YBCO. Initial dimension: 10.93 mm.
Heating rate: 20 OC/min. Top curve: heating cycle. Bottom curve:
cooling cycle. 91

4-1 Enthalpy of YBa2Cu307 as a function temperature. Charged species
not considered. 111

4-2 Energy distribution for the evaporation YBCO target using KrF
laser_ 113

5-1 YBCO film on LaAlO3 substrate produced by pulsed laser
deposition. Target heated to 700 oC. 129






5-2 YBCO film on MgO substrate produced by pulsed laser deposition.
Target at room temperature. 130

5-3 Surface morphology a heated target surface evaporated using KrF
excimer laser at 1.2 J/cm2. Magnification (a) 70 x, and (b) 1000 x. 132

5-4 YBCO film on LaA103 substrate produced by pulsed laser
deposition. (a) 10000x magnification (b) 30000x magnification.
Note the absence of large particles. 133

5-5 YBCO film on LaA103 substrate produced by pulsed laser
deposition using heated target at 700 oC. Note the absent of large
spherical particles. 134

5-6 YBCO film on LaA103 substrate produced by pulsed laser
deposition using heated target at 25 oC. Note the present of large
spherical particles. 135

5-7 Surface morphology a heated target surface evaporated using KrF
excimer laser at 2.5 J/cm2. Number pulses: 3000. Atmosphere:
100 mtorr 02. (a) magnification 50x and 250x and (b)
maginfication 600x and 3000x. Note the absence of surface
morphology. 137

5-8 Surface morphology a target surface evaporated using KrF excimer
laser at 2.5 J/cm2. Number pulses: 3000. Atmosphere: 100 mtorr
02. (a) magnification 50x and 250x and (b) maginfication 600x
and 3000x. Note the presence of surface morphology. 138

5-9 Surface morphology YBCO target surfaces evaporated using KrF
excimer laser at 2.5 J/cm2. Number pulses: 3000. Atmosphere:
100 mtorr 02. Magnification: 40x. (a) target at 25 oC, and (b)
target at 600 oC. 139

5-10 XRD YBCO target after heating at 600 oC for 2 hours. Loss of
oxygen can be noticed. 140

5-11 Typical temperature dependent surface resistivity. 141

5-12 Typical XRD of laser deposited thin film on MgO substate, showing
c-axis orientation. 142

6-1 High density superconductor pellet target. (a) nonpolished surface,
(b) polished surface. 152
6-2 Schematic diagram of the microfabrication stage used in the
evaporation YBCO target 153

6-3 YBCO target evaporated at fluence (a) 0.1 to 0.6 J/cm2, (b) 0.6 to
1.1 J/cm2, and (c) above 1.1 J/cm2. Corresponding to the
formation of pores, cones and melts surface morphology. 155







6-4

6-5


6-6



6-7


YBCO target evaporated using KrF laser.
number: 200 pulses. Fluence: 0.6 J/cm2.
Top view.

YBCO target evaporated using KrF laser.
number: 500 pulses. Fluence: 0.6 J/cm2.
(a) top view; (b) cross-sectional view.


YBCO target evaporated using KrF
number: 1000 pulses. Fluence: 0.6
(a) top view; (b) cross-sectional vie,


Atmosphere: air. Pulse
Repetition rate: 10 pps.
1

Atmosphere: air. Pulse
Repetition rate: 10 pps.


163


laser. Atmosphere: air. Pulse
J/cm2. Repetition rate: 10 pps.
w. 16


Cartoon for the evolution surface shielding due to formation yttrium
rich phase. (a) initial absorption; (b) formation of yttrium-rich phase;
(c) material shielding. 166

XRD YBCO (a) virgin material; (b) laser scanned at low fluence.
Note the presence of Y203 phase. 167

Experimental and calculated phase diagrams for the Y-Ba-Cu-O
system. 168

YBCO target evaporated using KrF laser at fluence between 0.6 to
1.1 J/cm2. Formation of cones surface morphology. Atmosphere:
air. Pulse number (a) 50 pulses; (b) 200 pulses; (c) 1000pulses. 170


YBCO target evaporated using KrF 1
number: 50 pulses. Fluence: 0.9 J/
(a) top view; (b) cross-sectional view


YBCO target evaporated using KrF laser.
number: 200 pulses. Fluence: 0.9 J/cm2.
Top view. Note the formation of "nipples".


aser. Atmosphere: air. Pulse
cm2. Repetition rate: 10 pps.
. 1


Atmosphere: air. Pulse
Repetition rate: 10 pps.
.. ... .17


YBCO target evaporated using KrF laser. Atmosphere: air. Pulse
number: 500 pulses. Fluence: 0.9 J/cm2. Repetition rate: 10 pps.
(a) top view; (b) cross-sectional view. 174


xvii


Etch rate YBCO as a function laser fluence. Inset graph is in linear
scale. 157

EDAX analysis YBCO target evaporated at fluence from 0.1 to 2.0
J/cm2. The fraction of yttrium is high at fluence less than 1 J/cm2. 158

YBCO target evaporated using KrF laser at fluence between 0.1 to
0.6 J/cm2. Formation of pores surface morphology. Atmosphere:
air. Pulse number (a) 50 pulses; (b) 200 pulses; (c) 1000pulses. 160

YBCO target evaporated using KrF laser. Atmosphere: air. Pulse
number: 50 pulses. Fluence: 0.6 J/cm2. Repetition rate: 10 pps.
(a) top view; (b) cross-sectional view. 161


6-10


6-11


6-12


6-13


6-14


6-15


6-16



6-17






6-18


YBCO target evaporated using KrF laser.
number: 50 pulses. Fluence: 2.0 J/cm2.
(a) top view; (b) cross-sectional view.


YBCO target evaporated using KrF laser.
number: 200 pulses. Fluence: 2.0 J/cm2.
(a) top view; (b) cross-sectional view.

YBCO target evaporated using KrF laser.
number: 500 pulses. Fluence: 2.0 J/cm2.
(a) top view; (b) cross-sectional view.


YBCO target evaporated using KrF laser.
number: 1000 pulses. Fluence: 2.0 J/cm2.
(a) top view; (b) cross-sectional view.


Atmosphere: air. Pulse
Repetition rate: 10 pps.
17


Atmosphere: air. Pulse
Repetition rate: 10 pps.
18
Atmosphere: air. Pulse
Repetition rate: 10 pps.
18


Atmosphere: air. Pulse
Repetition rate: 10 pps.


Polished YBCO surface. 184

Polished YBCO target evaporated using KrF laser. Fluence: 2.0
J/cm2. Atmosphere: air. Pulse number: (a) 1 pulse; (b) 2 pulses.
Note that at 2 pulses, some of the holes are filled. 185

Polished YBCO target evaporated using KrF laser. The same
surface area as 6-25. Fluence: 2.0 J/cm2. Atmosphere: air. Pulse
number: (a) 5 pulses; (b) 8 pulses. 187

Polished YBCO target evaporated using KrF laser. Fluence: 2.0
J/cm2. Atmosphere: air. Pulse number: (a) 10 pulses; (b) 20
pulses. 188

Polished YBCO target evaporated using KrF laser. The same
surface area as 6-27. Fluence: 2.0 J/cm2. Atmosphere: air. Pulse
number: (a) 50 pulses; (b) 80 pulses. 189

Polished YBCO target evaporated using KrF laser. Fluence: 2.0
J/cm2. Atmosphere: air. Pulse number: (a) 100 pulses; (b) 200
pulses. 190

Polished YBCO target evaporated using KrF laser. The same
surface area as 6-29. Fluence: 2.0 J/cm2. Atmosphere: air. Pulse
number: (a) 500 pulses; (b) 800 pulses. 191


xviii


YBCO target evaporated using KrF laser. Atmosphere: air. Pulse
number: 1000 pulses. Fluence: 0.9 J/cm2. Repetition rate: 10 pps.
(a) top view; (b) cross-sectional view. 175

YBCO target evaporated using KrF laser at fluence above 1.1 J/cm2.
Formation of melts surface morphology. Atmosphere: air. Pulse
number (a) 50 pulses; (b) 200 pulses; (c) 1000pulses. 177


6-19


6-20


6-21


6-22


6-23


6-24

6-25



6-26



6-27



6-28



6-29


6-30







6-31 Y2BaCuO5 target evaporated using KrF laser. Atmosphere: air.
Condition: (a) Virgin material; (b) 0.4 J/cm2,1000 pulses 193

6-32 Y2BaCuO5 target evaporated using KrF laser. Atmosphere: air.
Condition: (a) 0.7 J/cm2, 1000 pulses, 10 pps; (b) 1.1 J/cm2, 1000
pulses, 10 pps 195
6-33 Y2BaCuO5 target evaporated using KrF laser. Atmosphere: air.
Condition: (a) 2.1 J/cm2, 1000 pulses, 10 pps; (b) 3.5 J/cm2, 1000
pulses, 10 pps 196
6-34 Comparison between the etch rates of YBa2Cu307 and Y2BaCuO5
materials as a function fluence. 197
6-35 Etch depth Y2BaCuO5 as a function fluence and pulse number,
evaluated at fluences from 1.7 to 3.5 J/cm2. 198
6-36 Etch depth of YBa2Cu307 as a function fluence and pulse number. 199
6-37 Y203 target evaporated using KrF laser. Atmosphere: air. Fluence:
3.0 J/cm2. Pulse number: 200 pulses. 200
6-38 YBCO target evaporated using KrF laser. Atmosphere: air. Fluence:
0.5 J/cm2. Repetition rates: (a) 10 pps (b) 100 pps, 202
6-39 YBCO target evaporated using KrF laser. Atmosphere: air.
Fluence: 0.8 J/cm2. Repetition rates: (a) 5 pps; (b) 100 pps. Note
that surface (b) is smoother. 203
6-40 YBCO target evaporated using KrF laser. Atmosphere: air.
Fluence: 1.1 J/cm2. Pulse rates: (a) 5 pps; (b) 10 pps. 204
6-41 YBCO target evaporated using KrF laser. Atmosphere: air.
Fluence: 0.4 J/cm2 Size: 200 pm x 200 nm. Pulse number: 1000.
Repetition rate: 10 pps. Top view. 205
6-42 YBCO target evaporated using KrF laser. Atmosphere: air. Size:
200 gm x 200 gm. Pulse number: 1000. Repetition rate: 10 pps.
Fluence: (a) 0.9 J/cm2; (b) 1.4 J/cm2 206
6-43 YBCO target evaporated using KrF laser. Atmosphere: air.
Fluence: 2 J/cm2. Pulse number: 20. Repetion rate: Ipps. Target
density: (a) low density; (b) high density 209
6-44 YBCO target evaporated using KrF laser. Atmosphere: air.
Fluence: 2 J/cm2. Pulse number: 50. Repetition rate: Ipps. Target
density: (a) low density; (b) high density 210






6-45 YBCO target evaporated using KrF laser. Atmosphere: air.
Fluence: 2 J/cm2. Pulse number: 100. Repetition rate: Ipps.
Target density: (a) low density; (b) high density 211
6-46 LaAO03 target evaporated using KrF laser. Atmosphere: air. Pulse
number: 1000. Repetition rate: 10 pps. Fluence: (a) 0.6 J/cm2:(b)
0.8 J/cm2 213

6-47 LaA103 target evaporated using KrF laser. Atmosphere: air. Pulse
number: 1000. Repetition rate: 10 pps. Fluence: (a) 1.0 J/cm2; (b)
1.3 J/cm2 215
6-48 LaA103 target evaporated using KrF laser. Atmosphere: air.
Fluence: 2.1 J/cm2. Repetition rate: 10 pps. Pulse number: (a)
200; b) 1000 216

6-49 A1203 target evaporated using KrF laser. Atmosphere: air. Fluence:
15 J/cm2 217

6-50 Titanium target evaporated using KrF laser. Atmosphere: air. Pulse
number: 500 pulses. Repetition rate: 50 pps. Fluence: (a) 0.2
J/cm2; (b) 0.4 J/cm2 219

6-51 Titanium target evaporated using KrF laser. Atmosphere: air. Pulse
number: 500 pulses. Repetition rate: 50 pps. Fluence: (a) 0.8
J/cm2; (b) 1.6 J/cm2 220

6-52 Gold target evaporated using KrF laser. Atmosphere: air. Fluence:
2.0 J/cm2. Note the formation of surface ripples. 221

6-53 Polyimide Kapton target evaporated using KrF laser. Atmosphere:
air. Fluence: 25 mJ/cm2. The surface morphology showed the
presence of debris. 223

6-54 Cross sectional view YBCO target evaporated using KrF laser.
Fluence: 2.0 J/cm2. Pulse number: 1000. Note the formation of
melt on the crater wall. (a) cross sectional view of the crater; (b)
close up of the deposit on the left wall. 225

6-55 Cross sectional view YBCO target evaporated using KrF laser.
Fluence: 1.5 J/cm2. Pulse number: 500. Note the formation of
column on the crater bottom. (a) cross sectional view of the crater;
(b) close up of the deposit on the column. 226

6-56 Cross sectional view YBCO-gold powder target evaporated using
KrF laser. Notice the formation gold spheres at the tip of the
columns. 230

6-57 Cartoon of ultra thin surface crack modeL 233






6-58 Ray tracing model used in the development surface morphology. As
the pulse number increases, the etch depth becomes deeper and the
broader. 238

6-59 Surface absorbance as a function wavelength. Wavelength from
220 to 750 nm. 240

6-60 Surface reflectivity as a function of incident angles. 241

6-61 Evolution of YBCO surface morphology at fluence of 2J/cm2 as a
function of pulse number for both polished and non-polished
targets 242

7-1 Redeposited YBCO plume as a function fluence. 251

7-2 Redeposited plume area as a function of fluence. Linear dependency
observed for both 10 pps and 100 pps. 252

7-3 YBCO target evaporated using KrF laser. Pulse number: 500.
Repetition rates (top to bottom): 100, 10 and 5 pps. Fluences: (a)
1.7, 0.1, 0.2, 0.3, 0.4, and 0.5 J/cm2; (b) 0.5, 0.6, 0.7, 0.8, 0.9,
and 1.0 J/cm2. 253

7-4 YBCO target evaporated using KrF laser. Pulse number: 500.
Repetition rates (top to bottom): 100, 10 and 5 pps. Fluences: (a)
1.0, 1.1, 1.2, 1.3, 1.7 and 2.0 J/cm2; (b) 1.0, 1.1 and 1.2 J/cm2,
10 and 5 pps. 255

7-5 Redeposited plume as a function of fluence for different evaporation
sizes for YBCO target. 256

7-6 Titanium target evaporated using KrF laser. Fluence from 0.1 to 2.0
J/cm2. 257

8-1 Schematic diagram a differentially pumped mass spectrometer. 266

8-2 Detail the orifice design at the differentially pumped mass
spectrometer. Orifice diameter: 100 pm. Orifice length: 100 im. 267

8-3 Mass spectrometer (MID) as a function fluence. (a) with ionizer,
detecting both neutral and ionic species; (b) without ionizer,
detecting only ionic species. 269
8-4 A representative mass spectra for YBCO plume. Note the presence
of atomic species. 273











Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


STUDIES OF LASER-TARGET INTERACTIONS IN PULSED EXCIMER LASER
EVAPORATION OF SUPERCONDUCTING OXIDES AND OTHER METAL OXIDES

By

Stephens Sin-Tsun Th6

August 1993



Chairman : Professor Timothy J. Anderson
Major Department: Chemical Engineering

A detailed study of the interaction of a KrF beam with a Y-Ba-Cu-O high-

temperature superconductor target was performed to address two principal problems

associated with thin film deposition via pulsed laser evaporation: (1) formation of film

surface particulates and (2) development of irregular target surface morphology.

Thermodynamic chemical equilibrium and mass spectroscopic analysis were performed to

determine the composition of the plasma generated by laser irradiation of Y-Ba-Cu-O

during the formation and expansion stages.

The formation of film particulates was suppressed by maintaining the target bulk

temperature greater than 700 OC. Such target heating eliminated the formation of spherical

film particulates greater than one micron and suppressed the evolution of target surface

morphology (i.e., pores, cones, and melts).

The development of surface morphology entailing the distribution of laser power

via optical and conventional thermal analysis was investigated. A detailed fluence-

dependent optical analysis was used to describe the surface morphology evolution which


xxii












was found to be a strong function of laser fluence and pulse number. An overall energy

balance showed that the phase transformation consumed one-third of the absorbed laser

energy with the remainder converted to plasma kinetic energy. This kinetic energy was

manifested in the redeposited plume area formed during ambient pressure evaporation;

plume area was a linear function of fluence. The physical properties used in calculations

were determined using thermal methods (heat capacity, transition temperatures, and thermal

expansion coefficient) and laser flash technique (thermal diffusivity).

A thermodynamic equilibrium calculation (stoichiometric algorithm) was performed

on a high-temperature, high-pressure, adiabatically expanding system to determine the

chemical equilibrium as a function of temperature, pressure, and inert gas concentration.

Ion formation was favored at high temperatures, low pressures, and high inert gas

concentrations. Oxide formation was favored at high oxygen pressures and low

temperatures. Chemical characterization of the plasma plume using a differentially pumped

quadrupole mass spectrometer (P <30 mtorr) showed the presence of mainly neutral and

ionic atomic species. The result was consistent with the thermodynamic equilibrium

calculation.

For comparison, the effect of KrF laser irradiation on other materials, including

refractory ceramics, biocompatible materials, and polymer films, was also investigated.


xxiii











CHAPTER 1
INTRODUCTION

The recent discovery of high Tc superconductors (HTSC) has prompted

considerable interest in new fields of superconductor applications, including electronics.

However, this application requires smooth, high-quality film, and the quality of film

depends largely on the deposition technique. Pulsed laser deposition (PLD) has emerged

as a very versatile technique for thin film deposition. PLD is effective for depositing

multicomponent materials while maintaining the stoichiometry of the film with respect to

the laser target.' More than a hundred different materials 2 containing up to six elements,3

such as superconductors,4 semiconductors,5 metals,1 dielectrics,1 and bioceramics,6 have

been deposited using PLD. Some of the highest quality HTSC thin films have also been

produced by this technique.4

PLD is an experimentally simple, thin-film deposition technique, although the

detailed process mechanisms are complex and not well understood. In this deposition

process, photon energy is absorbed by target materials to produce volatile species that are

subsequently deposited as thin films on a substrate. The process phenomena, however, are

complex, involving absorption of the photon energy by the target, followed by the

conversion of photon energy into electronic, thermal, chemical, and mechanical energies,7

and finally evaporation, excitation, plasma free-expansion, and film deposition. Specific

subprocesses can involve near equilibrium mixtures as well as highly nonequilibrium

reacting systems. In a recent review, Hubler described PLD as a simple method, the

effectiveness of which no one exactly understands.3

Studies of laser-material interaction were pioneered by Ready in 1963 8 after the

invention of the laser in 1960.9 Early work on laser-assisted thin-film deposition was




2

initiated in 1965 by Smith and Turner,10 using a ruby laser to deposit semiconductors and

dielectrics. Although this work was conducted early in the development of the laser

deposition technique, two fundamental observations were made: 1) the interaction between

the laser beam and the target material produces a high-velocity plasma plume, and 2) the

stoichiometry of the deposited film is preserved. Limited by the availability of high-energy

lasers and suitable materials, this deposition technique was largely ignored until the

invention of the high-energy pulsed excimer laser around 1975 and the recent discovery of

HTSC."1 Bellcore/Rutgers University 4 pioneered the use of pulsed laser deposition of

HTSC thin films by successfully making the first high-quality YBa2Cu307 thin film in

1987. Presently, PLD is one of the most rapidly developing thin-film deposition

techniques for growing thin films of advanced materials.

PLD has several positive attributes. Its intrinsic capacity to preserve stoichiometry

during deposition of multicomponent materials makes it especially suited for materials that

evaporate incongruently.5 In addition, the formation of a highly directional plume,12

perpendicular to the target surface, makes PLD very efficient for transferring target

materials. The highly energetic plasma plume imparts significant species surface mobility

on the substrate surface,3 permitting deposition at a lower substrate temperature.' 13

Because the evaporation is induced by an external laser source, in contrast to a sputtering

process that uses accelerated background gases, the PLD chamber pressure is decoupled

from the evaporation mechanism and can be independently adjusted.14 Moreover,

multilayered structures can be grown by simply switching the target materials during the

deposition process.15

Compared to the widely used sputtering process, PLD is still an immature

technology, with the underlying physics and chemistry still largely unknown. As a result,

several practical issues need to be addressed before PLD can become a dependable

deposition technique. These issues include the generation of micron-sized particles, which

lead to poor film surface morphology; the formation of irregular target surface







morphologies, which affect the deposition efficiency; and the production of a limited plume

cross section, which restricts the use of PLD for small areas. Solving these problems is

crucial to the broad adoption of this process by the electronic industry, which traditionally

demands high process efficiency and adaptability.

This dissertation addresses some of the key issues associated with the deposition of

Y-Ba-Cu-O high-temperature, superconducting thin films. In particular, attention is given

to understanding the mechanism by which different target morphologies are developed,

understanding the composition of the plasma plume, and investigating methods to reduce

the formation of particulates. This introduction presents an overview of HTSC materials

and their applications, discusses the general features of the PLD process and provides a

perspective of the work included in this dissertation.

1.1 Overview of High Tc Superconductor Thin Film

This section deals with the basic properties of superconductors and their

applications, especially as they pertain to thin films used for microwave devices. The

criteria for high-quality thin films, film characterization techniques, and various thin-film

deposition techniques are discussed.


1.1.1 Status and Applications


Superconductor is an ideal material in the fabrication of electronic components

because it exhibits zero-resistance properties. To understand the current status of

superconductors and their applications in research, the use of superconductors in passive

and active microwave devices, as well as integrated systems, is examined.

1.1.1.1 Superconductor and its properties

One of the most important characteristics of a superconductor is its ability to carry

current without any resistance. This departure from normal electrical behavior allows

electrons to be conducted through a superconductor without energy loss or Joule heating.







Superconductivity was discovered and investigated by Onnes in 1911 in his study of solid

mercury (Hg) at low temperature.16 The abrupt transition from an ordinary conductor to a

superconductor occurs at the critical temperature, Tc. In the case of Hg, this transition

occurs at 4 K, the boiling point of helium at zero field. Two other important characteristics

of this transition are the critical current density, Jc, and the critical magnetic field, He,

which specify the upper limit of the applied current and magnetic field for superconducting

behavior. A superconductor is not only a perfect electrical conductor, it also is a perfect

diamagnet, with the capability to expel a magnetic field. This characteristic, which renders

a superconductor capable of magnetic levitation, is known as the Meissner effect. If a

magnetic field greater than He is applied, however, it will abruptly penetrate the

superconductor material and destroy the superconductivity. In a similar manner,

superconductivity will abruptly vanish beyond the critical current level, Jc.

There are two classes of superconductors based on the transition temperature range:

low- and high-temperature superconductors. A low-temperature superconductor becomes a

superconductor in the liquid helium temperature range. Common examples are niobium-

based metal superconductors, such as pure niobium with a critical transition temperature,

Tc, of 9 K. A high-temperature superconductor, on the other hand, becomes a

superconductor in the liquid nitrogen temperature range; rare-earth oxide-superconductors

are examples. Based on operating cost, high-temperature superconductors are more

economical than metal superconductors because the required cooling can be supplied by

liquid nitrogen instead of the more expensive liquid helium, which reduces the cooling cost

by four orders of magnitude.16

In spite of their higher operating cost, low-temperature superconductors have

broader applications than high-temperature superconductors. Low-temperature

superconductors, because they are more malleable, can be more easily shaped into useful

forms such as superconducting magnetic coils and cables. Applications include the

superconducting supercollider, the magnetically levitated train, and magnetic resonant




5

imaging, as well as applications in superconducting circuits. In contrast, the high-

temperature superconductor is brittle, currently restricting applications to superconducting

circuits.

High-temperature superconductors are ceramic materials with an oxygen-depleted,

layered perovskite structure. A perovskite structure consists of a framework of a cubic

structure with three distinct elements with a chemical formula ABO3, where A and B are

cations, which occupy the center and the covers of the cube, respectively; the oxygen

anions occupy the centers of the edges of the cube.

The most common high Tc superconductor is YBa2Cu307-x, first reported by Chu

et al.11 This superconductor is also known as the 123 compound or YBCO

superconductor with a Tc value near 95 K (Figure 1-1).17 The general formula for this

class of material is REBa2Cu307-x, where RE comprises the rare earths Y, La, Nd, Sm,

Eu, Gd, Dy, Ho, Er, Tm, Yb, and Lu, and x is between 0.095 and 0.3.18

In YBa2Cu307, Y and Ba atoms occupy the A cation sites; whereas, Cu ions

occupy the B cation sites (Figure 1-1). It is believed that the current is transported in the

copper-oxygen plane of this unit cell, which is parallel to the substrate in c-axis oriented

film.16 Vertical-device structure applications require a-axis oriented film with the Cu-O

plane normal to the substrate.

Because of the differences in the relative sizes of the ions, a variety of distortions

from the cubic structure can take place. As the result, a YBaCuO superconductor has two

crystalline forms: orthorhombic and tetragonal. The orthorhombic state is the

superconducting state with the oxygen stoichiometric in the range of 6.5 to 7. The

tetragonal state has an oxygen stoichiometry less than 6.5 and a Tc value of 66 K. The unit

cell dimensions of orthorhombic YBaCuO superconductor are a=3.8 <, b=3.9 <, and

c=13.0 or 27.2 <.16 An excellent review of the properties and processing of this material

is available.19




6













c





0.41nm Y
O Ba 5 I 200
e *Cu 4'" *
SIn --n o -
0.34 nm .150
0 0 3 3 3 0
'o f-" 100 'o
0.41 nm
50 ..4
S 0.38nm b5
S0.39nm 0 100 200 300
T (K)
(0) (b)















Figure 1-1 YBa2Cu307 superconductor (a) atomic structure; (b) temperature
dependent resistivity. Reference 17.







Other technologically important classes of superconductors include the thallium-

based compounds, e.g., Tl-Ba-Ca-Cu-O (Tc=125 K) and Tl-Pb-Sr-Ca-Cu-O (Tc=125 K),

and bismuth-based compounds, e.g., Bi-Sr-Ca-Cu-O (Tc= 10 K). The general formula of

these compounds is (A"'O)xA2"Can-lCunO2n+2, where Am = thallium, bismuth,

bismuth+lead, bismuth+thallium, RE or thallium+lead; Al= barium or strontium; and n =

number of copper-oxygen (CuO2)2- layers.18 Because these superconductors have more

elements than YBCO, more secondary phases are present, which makes it difficult to

produce high-quality and stoichiometric films. A problem with thallium-based films is that

the high vapor pressure of thallium makes it difficult to control the stoichiometry. An

advantage of this material, however, is that the grain boundaries have less influence on the

surface resistivity than in YBCO films.

The qualities of high-temperature superconducting thin films make them suitable for
a variety of applications including passive microwave devices (e.g., passive interconnects)

and active microwave devices (e.g., quantum interference devices and tunneling junction

switches).20 It is important to consider the potential dynamic application in determining

how the film should be optimized and in understanding the quality of film to be made.

These device applications are discussed in detail below.

1.1.1.2 Passive microwave devices

Passive devices are constructed from a patterned, single-layer superconducting

film. Passive microwave devices change the input signals in a linear fashion,21 that is, the

output varies linearly with the input. Applications of this type of device are microwave

filters, resonators, switches, and delay lines. A schematic view of a stripline resonator is

shown in Figure 1-2.21 With these devices, only the surface properties of the

superconducting film are important, mainly, the superconducting zero resistance property.

The device is primarily used as the signal transfer media in which the signals travel at the

surface.




























TRANSMISSION
LINE


CENTER
CONDUCTOR


\ GROUND
PLANES


LOW-LOSS
DIELECTRIC

STRIPLINE CROSS-SECTION
i


TOP VIEW


Schematic view a stripline resonator. (a) cross section showing
conductor and dielectric; (b) top view. Reference 21.


GAP


Figure 1-2







Devices fabricated of superconductor material are superior to those that use a metal

(e.g., copper) since the superconductor surface resistance value is a few orders of

magnitude lower. The microwave surface resistance of copper at 77 K and 10 GHz is
typically 10 mn, whereas for an HTSC, the value is 200 g2.22

The main application for passive microwave devices is in signal processors, which

accept signals from analog sources and extract the essential information by improving the

signal-to-noise ratio. Radar is an example of signal processing used to produce a map of

target intensity as a function of range and velocity.

1.1.1.3 Active microwave devices

Active devices have more complex structures than passive devices and process

signals nonlinearly.21 The best example is the Josephson junction. The intrinsic speed of

superconductive electronics comes from the rapid switching ability of Josephson junctions.

A cross section of a circuit containing Josephson junctions is shown in Figure 1-3.23

Constructing a Josephson junction involves complex, multilayered superconductor

film formation with critical control of the interface and thickness. A typical Josephson

junction consists of two layers of superconductor separated by a thin insulator. The

thickness of the insulator is of the same order as the coherence length; that is, the thickness

for "tunneling" to take place, which, depending on the crystal orientation, is from 3 to 35 A

for YBCO material.24

In principle, Josephson switches use a tunnel junction that operates between the

superconducting state and the voltage state, corresponding to the logical "0" and "1" states.

The device switching time from the zero-resistant state to the finite-voltage state is in the

range of picoseconds. This rapid switching allows an electronic signal to be controlled in

the same time frame, which is much faster than current computers with time constants on

the order of nanoseconds. Since light travels a foot in a nanosecond, the future fast

superconducting computer has to be small, about the size of a calculator, to minimize

distance; otherwise the fast switching time is useless.




















HTS Josephson
Junctions


crossovers
Normal metal
resistor


Superconducting
interconnect
\


YBCO

S Junction metal


I i Epitaxial
insulator
Resistor metal


Schematic view a HTSC Josephson Junction circuit. Reference 23.


Figure 1-3







1.1.1.4 Integrated systems

Beyond active and passive devices, there are integrated systems, which consist of

several devices. Examples are superconducting quantum interference devices (SQUIDs),

Josephson arrays, microwave detectors, digital signal processors, and analog signal

processors.20 A cross section of an integrated SQUID is shown in Figure 1-4.23

A SQUID, in principle, is a flux-to-voltage transducer, which was developed at

Ford Motor in 1960. It combines two physical phenomena: flux quantization and

Josephson tunneling. Fundamentally, it is a superconductor ring that acts as a storage

device for magnetic fields, and was intentionally made to contain weak links of Josephson

junctions. Its main function is to serve as a magnetic flux gate. A superconductor ring will

oppose an applied magnetic field by generating superconducting current. If the field is

large, the current will also be large and beyond the Josephson junction capacity, causing

the junction to fail. This breaks continuity and allows more "quanta" of magnet to pass. If

a known and variable current is applied to the ring to balance this effect, the field can be

detected using a feedback method.

Since SQUIDs have quantum-level sensitivity to magnetic flux, they are used as

ultra-sensitive magnetic-field detectors, capable of detecting fields less than a billionth of

the naturally occurring earth field. Applications of SQUIDs include passive devices for

medical diagnostics, mineral surveying, earth mapping, submarine detection, relative

motion detection, and scientific instruments; and active devices for low-noise, high-

frequency mixers, high-speed oscilloscopes, and A/D converters.

The development of the Josephson junction and SQUIDs is currently limited by the

lack of high-quality thin films. The film must be smooth and uniform at a length scale less

than the smallest circuit feature in the order of a few nanometers; so far, no film deposition

technique can meet this criterion. Fabrication of the Josephson junction is further

complicated by the requirement of high substrate temperature during deposition, which
















(a)


1000 A SrTiO3
100 A PBCO
inn A (r'n,


Cross sectional view a integrated SQUID magnetometer (a) on
LaA103 substrate; (b) on YSZ substrate. Reference 23.


Figure 1-4


(b)







makes it almost impossible to fabricate two superconducting layers sandwiched through an

insulating barrier without encountering interdiffusion problems.

1.1.2 HTSC Thin Films

The quality of the superconducting film is the key to a high-performance microwave

device. This section deals with the selection and testing criteria for an ideal thin film,

selection of the substrate, and the technological challenges in producing the

superconducting thin film.

1.1.2.1 Ideal film properties

Ideally, a high-quality HTSC thin film possesses good stoichiometry, high phase

purity, perfect crystalline structure, high Tc, small transition width, high Jc, low surface

resistance, and smooth surface morphology. These are, in fact, the characteristics of a

single, perfect superconductor crystal. In reality, however, films only have some of these

properties. At a minimum, a useful high Tc film should be epitaxially grown on a lattice-

matched substrate with the correct crystalline structure and be relatively smooth; i.e., have

only a few boulders and pinhole defects. It is important to point out that high Tc film does

not necessarily have a high Jc, since only one direct superconductive path is required to

give good Tc, in contrast to Jc, which is determined by the bulk surface resistance. The

presence of in-plane misorientation and grain boundaries strongly affects the film's high-

frequency surface resistance, which is important in microwave applications.

1.1.2.2 Characterization techniques

Many analytical techniques are available to evaluate the performance and quality of

thin films. They are specifically used to determine the film composition, Tc, Jc, and crystal

orientation. The techniques used in this dissertation are briefly described below. This

thesis focuses on the laser target and plasma plume interaction, and the deposited film

provides a record of this interaction.







Composition.

Rutherford backscattering (RBS), used to determine the thin-film composition, is

based on the elastic collisions between accelerated 2He4 atoms and the atom constituents in

the film. Atomic weights and distributions of elements in the film can be determined from

the measured momentum of the reflected helium atoms momentum.

Electron micro probe analysis (EMPA), which was used to determine composition,

was based on the x-ray produced when the outer-shell electrons drop to the inner-shell

vacancies. The element can be identified from the x-ray energy released.

Auger electron spectroscopy (AES), used to determine composition, was based on

the Auger electron produced when electrons from the outer shell drop to the vacancies in

the inner shell. An Auger electron has a kinetic energy characteristic of the host atom.


Eddy current measurement was used to determine the Tc onset and transition width.

It works by detecting the inductive current generated by the film's superconducting

magnetic fields. As the film undergoes a zero resistance transition, the inductive current

sharply increases.


Four-point DC current probes and AC surface current measurements were used to

determine the Jc value and surface resistance of the film. DC current measurements are

simply contact probe resistance measurements, while AC current measurement are used to

determine the surface inductance.

Crystal structure.

X-ray diffraction (XRD) was used to determine the phase purity and crystal

orientation of the film. Based on the diffracted x-ray beam by the crystal lattice, the crystal

spacing and coherence of each phase can be determined.







Surface morphology

The scanning electron microscope (SEM) is used to determine the film surface

morphology. The surface image can be constructed from the reflected electrons. Most

units are equipped with EDAX (energy dispersive x-ray analysis), which determines

surface composition based on the electron kinetic energy.

1.1.2.3 Substrates

Success in making high-quality HTSC thin film depends on the substrate chosen.

The substrate influences the film quality through the following properties: thermal

expansion mismatch, which can cause cracks in the film; interface reaction, which changes

the film composition; and, most importantly, substrate lattice mismatch, which affects the

film crystallographic orientation.25 The ideal substrate for depositing YBCO film is lattice-

matched with the perovskite structure (e.g., SrTiO3, LaA103, and LiNbO3) and has a

reasonable thermal expansion match and minimum interface reactivity.

In addition to the lattice matching, thermal stability, and reactivity properties, the

substrate dielectric characteristic should be carefully considered for microwave

applications. Many of the available substrates (such as LiNbO3, MgO, ZrO2, LaA103, and

SrTiO3) have some, but not all, of these desired properties. For example, LiNbO3 has

good dielectric and crystalline matches, but is very reactive, permitting Ba and Cu

migration through the interface. MgO and ZrO2 have ideal dielectric characteristic for

microwave applications. They are chemically inert with superconducting material, but

show poor crystal matching properties. LaAlO3 and SrTiO3, on the other hand, are perfect

in crystal match, but have high dielectric constants, which are poor for microwave circuits.

Similarly, SrTiO3 has an extremely high dielectric constant, which results in considerable

high-frequency losses in microwave applications. LaGaO3 has the lowest dielectric

constant and a lattice constant comparable to YBaCuO (orthorhombic, a=3.820 A, b=3.892

A, c/3=3.896 A). It is, however, susceptible to a phase transition at 140 C, which is

below the temperature range required for film annealing.26







Other choices for substrate are glass, quartz, sapphire, Sr2TiO4, BaTiO3, and

BaF2. Recent technology combines the characteristics of two or more materials in forming

buffer layers between a substrate and the high Tc film. Two examples are SrTiO3-buffered

MgO and cerium-oxide buffered zirconia.

1.1.2.4 Technological challenges

A quality HTSC is defined as having good crystalline structure and a smooth

surface. The crystalline structure determines the thin film performance, while the surface

smoothness determines the circuit resolution. High deposition temperatures are required to

enhance surface mobility during film growth, which results in proper crystalline structure.

However, imperfections arise in the film quality since superconductor materials are

inherently reactive 27 and the crystalline structure is sensitive to oxygen stoichiometry.

High temperatures exacerbate these problems, especially in multilayer thin films.

The above constraints pose many technological challenges in producing good

HTSC thin film. Some of these challenges are directly related to deposition techniques

used to produce thin film. Various deposition techniques and their merits are discussed

below.


1.1.3 Comparison of Techniques for Deposition of YBCO


The deposition of epitaxial thin films has been demonstrated using a number of

techniques,28,29 including electron-beam evaporation, sputtering, chemical vapor

deposition, laser deposition, spray pyrolysis, molecular-beam epitaxy, and activated

reactive evaporation. Although pulsed-laser deposition was the only technique used in this

research, a brief comparison of the four most common techniques is presented below. An

expanded discussion of pulsed-laser deposition follows this comparison.

1.1.3.1 Electron-beam evaporation

This was the first technique used for depositing HTSC thin films. This process

involves the evaporation of a solid target via electron-beam heating. Since the 123 material







does not melt congruently, separate sources are used. Early work was performed in

laboratories at IBM, Stanford, Cornell, and later at Kyoto University and AT&T. In order

to achieve a stoichiometric film, it is necessary to control the fluxes from the individual

sources, which can be elemental or compound. Some of the best films have been made by

the three-electron-gun evaporation technique. This technique is the most sophisticated in

terms of complexity and expense, and it is not favored when a large number of elements

must be deposited.

1.1.3.2 Sputtering

Sputtering works by the momentum transfer between the sputtering ions and the

atoms or molecules in the target. There are several types of sputtering systems, including

diode source, conventional magnetron, ion beam, and novel cathodes. The oxides can be

sputtered from multiple or single sources. This method was developed by Stanford,

Westinghouse, and Argonne. TRW has developed a three-target technique. The target

may be in a machined form or a powder. The composition of the film is altered by the

negative ion bombardment on the deposited film during the deposition. Off-stoichiometry

targets, correct geometry, or high pressure are used to correct the problem. This technique

is widely used for commercial applications and has no scale-up problem. Thin films as

large as several inches in diameter have been deposited. The drawback of this technique is

the slow deposition rate.

1.1.3.3 Metallorganic chemical vapor deposition (MOCVD)

This method is capable of epitaxial growth of multielement crystalline materials on

large-area substrates and is widely used to deposit compound semiconductors. This

technique uses volatile metallorganic precursors to transport the elements. The challenge is

to identify, produce, and handle precursors with suitable properties, especially for the

transport of barium, since volatile barium precursors are limited. The chemical complexity

of this method and the limited selection of volatile precursors have caused the slow

development of this technique.







1.1.3.4 Pulsed-laser deposition (PLD)

The pulsed-laser deposition technique utilizes photon energy to evaporate a target

material at low pressures and to deposit the resulting plasma as a thin film. Since the

irradiated surface temperature can reach several thousand degrees, a multicomponent target

can be evaporated stoichiometrically. Various laser sources can be used, including those

that emit in the far IR wavelength (e.g., a C02 laser) and in the near UV wavelength (e.g.,

the KrF excimer laser). This technique was broadly applied to HTSC deposition by

Bellcore and Rutgers University. Although the deposition rate is relatively high, it is

limited by the optical properties of the evaporated target material and the formation of

particulates on the film.

1.2 Overview of Pulsed-Laser Evaporation

Pulsed-laser deposition can be adapted for depositing a variety of materials,

including the high-temperature superconductors. It is the technique used in this study, and

a broad overview of the hardware, deposition parameters, laser-target interactions, and

current challenges is presented below. Particular attention is given to the formation of film

particulates and the development of target morphology for deposition of YBCO.

1.2.1 Hardware Configurations


Compared to other thin-film deposition techniques, a PLD growth system is a

relatively simple experimental design, with minimum hardware requirements inside the

vacuum chamber.30 A PLD system has two main components, a vacuum chamber and a

laser system.

A schematic diagram of the system used in this study is shown in Figure 1-5.

The basic vacuum chamber is configured with a target holder, a substrate holder, a

gas nozzle, and a glass window. The target holder is generally designed to allow rotation

in order to present new target areas to the laser during the evaporation. The substrate is























LENS


TYPICAL OPERATING CONDMONS



FLUENCE : -3 J/cm2

PRESSURE : 0.1 0.2 orr

SUBSTRATETEMP : 600.700C


PHOTON ABSORPTION


( 0 20 ns)


PHASE TRANSFORMATION
PHOT O1NIZATION
( 5 ps 20 ns )


EVAPORATION
CLUSTER EJECTION
( I 10 us )


Schematic diagram a pulsed laser deposition chamber. Bottom view
shows zones of laser-target interactions.


Figure 1-5







heated (2 700 OC) to allow optimum crystalline growth. The nozzle, placed at a location

away from the pump in the chamber, is used to establish a gas ambient, typically oxygen,

during deposition. The background pressure is typically adjusted in the range of 100 to

200 mTorr, requiring a flow rate in the range 50 to 100 sccm. For optional plasma in situ

monitoring, a PLD growth system can be equipped with a variety of diagnostic instruments

such as a mass spectrometer, optical spectrometer, or ion probe.

The laser system consists of a high-power laser, coupled to an appropriate beam

delivery system. Lasers are operated in continuous and pulsed modes. A continuous laser

produces a steady output of laser power. Pulsed laser, on the other hand, discharges the

laser energy periodically at high intensity in a short duration. Various types of lasers have

been used, including short wavelength UV pulsed excimer lasers, e.g., ArF (193 nm), KrF
(248 nm), XeCI (308 nm), and long wavelength IR lasers, e.g., CO2 laser (10.6 ltr),

Nd:YAG lasers, both pulsed (1.064 Im), and Q-switched (532, 355 nm). The peak power

of these lasers can be as high as 108 W/cm2, with a pulse width that varies between

nanoseconds and milliseconds. The selection of a particular laser depends on the material

to be deposited. An excimer laser, having a high absorption coefficient, requires a lower

evaporation threshold; however, the repetition rate is low, typically less than 250 pulse per

second (pps). In contrast, a Nd:YAG laser has a higher evaporation threshold because the

absorption coefficient is low; however, the repetition rate can be as high as 2000 pps 31

Both an excimer laser and a Nd:YAG laser are capable of producing high-performance

films, although it is argued that the longer wavelength lasers may be better since the higher

penetration depth produces a deeper molten layer. Thus, the material removal is dominated

by nonequilibrium molten droplets, instead of equilibrium vapor.31 32

1.2.2 Operational Procedures

The two key sets of design and operational parameters in PLD are associated with

the laser and the deposition chamber. The parameters associated with the laser operation







are external and consist of laser wavelength, laser fluence, repetition rate, number of pulses

and the incident angle. The parameters associated with the deposition chamber include the

deposition geometry (e.g., the distance and orientation between the target and the

substrate), and the deposition environment (e.g., the ambient gas species, background

pressure, substrate temperature, and the annealing cycle). The results from several

different laboratories indicate that similar operational parameters will yield high-quality

YBa2Cu307 films 4. 33-36 The optimal values of these operational parameters have been

largely determined by empirical methods.

A typical deposition run begins with the placement of a pellet of YBCO

superconductor into a target holder inside the vacuum chamber. The pellet is produced

using a sintering technique that is discussed in Chapter 3. The pellet mounted in the holder

rotates at speeds up to 10 rpm to ensure that the laser beam encounters a new area on

successive evaporations. The pulsed laser beam is directed into the chamber through a

quartz or fused silica window and focused onto the target surface. Typical laser pulse

energy densities varies from 1 to 3 J/cm2, with repetition rates of 1 to 10 pps and incident

angles from 30 to approaching 900 from the target normal. The target area impacted by the

laser beam produces a brilliant, elongated plasma plume, which is then deposited on a

substrate located at 1 to 5 cm from the target surface.4 The optimum distance for film

deposition depends on the background oxygen pressure.37 Though the typical target etch

rates are 70 nm per pulse, deposition rates are on the order of 1 to 4 A/pulse due to the

large ratio of substrate to ablated area.4 The deposited film has a thickness variation of

about 20% across the substrate of an area of 0.25 cm2.4 In early studies, the substrate
temperature was maintained at a low value (= 400 oC), and the chamber pressure was

regulated at low value (= 10-5 Torr). In this high-vacuum process, the film is in an

amorphous state and requires annealing in oxygen at a temperature over 800 OC. Because

of the susceptibility of this amorphous film to moisture uptake during transfer, the

procedure was improved by increasing the background pressure to the range of 100 to 200







mTorr of oxygen to maintain the correct film oxygen content. Following film deposition,

the film is annealed by increasing the oxygen chamber pressure close to atmospheric

pressure, while cooling the substrate temperature at a prescribed rate.38 39 With these

growth conditions, the resulting YBa2Cu307 film is black, shiny, and superconducting.

The substrate temperature and the oxygen pressure have been found to strongly

influence the superconducting properties of the films. It is known that the activation energy

required for in-diffusion oxygen increases from 0.5 to 1.5 eV for YBa2Cu306.7 and

YBa2Cu307 respectively,40 therefore the higher deposition temperature is preferred in

obtaining superconducting film. PLD films that are deposited at room temperature are

insulating, yellow-brown, and exhibit poor adhesion to the substrate; subsequent annealing

does not improve the conductivity.4 In contrast, heating the substrate to 450 oC during

deposition produces a shiny, brown film with good adhesion properties. Upon annealing
in oxygen at elevated temperature (=900 oC) and then cooling slowly, the film becomes

superconducting. This film, however, is rough and phase-segregated. In situ growth

eliminates the need for this annealing procedure; therefore, the phase-segregation is

minimized. At low deposition temperatures, films are typically a-axis orientated,41 while
film deposited at optimal conditions (=700 to 800 OC in 100 to 200 mTorr oxygen) are

oriented along the c-axis.

1.2.3 Typical Results

The YBCO films deposited by laser deposition have been measured to have Tc

values as high as 95 K 4 and Jc values as high as 5x106 A/cm2 at 77 K and 4x107 A/cm2 at
4 K. The surface resistances at 10 GHz have been measured to be as low as 10 p.ohm at 4

K and below 300 tohm at 77 K.

XRD measurements have shown that film performance depends on the substrate-

film crystalline match. If the film is deposited on a perovskite substrate, such as (100)

SrTiO3, XRD has shown that the film has a high degree of c-axis orientation. On the other




23

hand, a film that is deposited on a cubic sapphire 4 has fewer degrees of orientation. This

mismatch is reflected in the value of Tc with measured values of 95 K on (100) SrTiO3 4

and 75 K on sapphire.4

The critical grain size required for high-quality superconducting properties has been

found to depend on the substrate selection. Typical grain sizes, as determined by SEM, is

500 nm on SrTiO3 and 150 nm on A1203.4 Small particles are always observed on the film

surface with diameter of 0.5 to 2 gtm as detected by SEM. These particulates are seen

regardless the type of laser used.

Laser fluence has been found to affect the film composition, especially if a laser

other than an excimer laser is used. At low fluence (0.6-0.8 J/cm2), Y-deficient films are

obtained. At medium fluence (1-2 J/cm2), excess Y and Ba and deficient Cu films are

obtained. At high fluence (> 2 J/cm2), excess Ba and deficient Cu films are obtained.33

Venkatesan reported that stoichiometric films could be produced using a KrF excimer laser

at fluences greater than 0.9 J/cm2 at an incident angle greater relative to the target normal

than 200. With an XeCI laser, improved stoichiometry was found at a higher angle and at

an energy density of 4 J/cm2.

Another factor affecting film quality is reaction with the substrate,27 especially at

high deposition or annealing temperature. Films grown on sapphire and annealed at 800

OC gave a Ba-rich interfacial layer and, consequently, a Y-rich remaining layer.42 These

films had poor superconductivity characteristics, although a Rutherford backscattering

showed that these films were highly stoichiometric with composition within 1% of the

target.


1.2.4 Overview of Laser-Target Interactions


The sequential processes involved in laser evaporation are as follows. The photon
beam is absorbed in the outer layer of the target (o = 105 cm-1 at 248 nm),43 causing the

surface temperature to increase, which produces surface melting and evaporation (Figure







1-5). The rapid evaporation can develop a recoil pressure, which expels the molten pool

and produces droplet ejection. The plasma, therefore, consists of both liquid and vapor

phases. If the pulse duration is sufficiently long, absorption of the photon by the plasma

becomes more important. The solid-liquid-vapor-phase transformations take place in

several picoseconds, whereas the plasma ejection process takes place in several

nanoseconds, possibly because it is limited by Knudsen flow.

Model development can be simplified by dividing the process into two steps based

on the characteristic time constants. The first step involves the absorption of photons by

the target, causing the target to heat, melt, and ionize while spatially localized in the

picoseconds. The second step is the expansion of the resulting high-temperature, high-

pressure plasma into the free space during the subsequent several microseconds.

1.2.4.1 Photon absorption and phase transformation

Photons are absorbed by the electrons and phonons in the target lattice, by the

molten carrier through surface absorption, and by the emitted plasma. The initial stage of

the evaporation process is surface heating due to the absorption of the laser energy in a

short period of time. During the evaporation of the ceramic target, the laser power density

can be as high as 107 to 108W/cm2. The surface temperature is determined by the

wavelength and flux of the incoming radiation, the thermal diffusivity, heat capacity, and

the absorption coefficient of the target. In general, a one-dimensional heat transfer model

involving surface melting is used."

In a condition suitable for material removal, the surface temperature of the ceramic

must exceed the melting point. For example, with YBCO this corresponds to a threshold

energy density of 0.11 J/cm2 at wavelength of 248 nm (KrF laser).43 The corresponding

etch rate varies logarithmically with energy density with a reported inverse absorption

length of 2.3x105/cm or a penetration depth of 500 A.43 As evaporation progresses,

texturing of the target surface is taking place, which lowers the evaporation rate.







1.2.4.2 Plasma expansion

The shape of the expanding plume is characterized by a high order cosine

distribution function, as high as twelfth order. In comparison, a normal Langmuir or

Knudsen evaporation exhibits a first order cosine function.45 It has been suggested that the

cosine distribution function in pulse-laser deposition contains both the high order and the

first cosine functions due to the contributions of both the high-velocity and the low-velocity

components of the plasma. In addition to the total mass distribution, a composition

variation is also observed.46 This composition variation is negligible for the center portion

of the plume in the high cosine region, but becomes nonstoichiometric in the outer region.

The plasma velocity was measured to be in the order of 106 cm/sec.12

Various techniques have been used in an attempt to analyze the composition of the

plasma phase, including emission, fluorescence, and absorption spectroscopy. Although

widely used, emission spectroscopy does not give a representative composition of the

plasma. First ,this technique only detects species that have detectable quantum yields for

fluorescence, and second, not all the species in the plasma fluoresce. Another method,

which more accurately detects the plasma phase, is mass spectroscopy.

Mass spectroscopy has revealed that the plasma contains mostly neutral and charged

atomic species.47-51 Although most results showed that the plasma consists mainly of

atomic species, high molecular weight clusters have also been detected under enhanced

conditions with secondary radiation. Quantitative determination of the ion content using an

ion probe has shown that charged species constitute less than 10 percent of the total

species,52 with a minimum charge concentration at fluence of 3 J/cm2. At low fluence, the

mechanism of material ejection is dominated by electronic excitation of the solid, producing

mainly charged species.47








1.2.5 Some Outstanding Problems for PLD Technique

There are several practical issues that still need to be addressed to make PLD a

reliable technique. The main problems are the generation of micron-sized boulders and the

development of target texturing, which affect the deposited film quality and deposition

efficiency. Furthermore, PLD is currently used only in laboratory-scale and has not been

developed fully to process scale. Lastly, a problem common to all deposition techniques is

the required high deposition temperature, which causes substrate-film interface reaction.

1.2.5.1 Boulder generation

The main drawback of the PLD technique is the production of boulders or

particulates on the substrate. It has been suggested that the boulder formation is caused by

surface texturing, plume condensation, or subsurface explosions at the target, which

produces fragments that transport to the substrate. Targets that have been polished smooth

produce fewer boulders;53 however, it is impractical to polish the target surface repeatedly

during the deposition, once the target texturing takes place. Several schemes have been

used to destroy the particles in mid-air, such as secondary lasers and plasma rings, but

particle densities on the substrate remain greater than 104/cm2. Other schemes used

mechanical filters, which selectively remove the low-velocity components, based on the

concept that the larger mass particles have a lower velocity than the lower mass plasma.

1.2.5.2 Target stoichiometry and texturing

Since laser evaporation is a photon-activated evaporation process, the nature of the

evaporation depends on the optical characteristics of the target surface. As the target

surface becomes highly textured, the photon efficiency for material removal is affected by

factors such as surface reflectivity, surface scattering, and absorption nonuniformity. As a

result, the surface texturing leads to a decrease in the deposition rate.53 In addition, the

plasma direction changes towards the direction of the incident laser beam as the target

becomes more morphologically uneven. Several techniques have been tried to limit the







texturing problem, including rotating the target to increase the evaporated surface area. The

fundamental problem, however, remains unsolved.

1.2.5.3 Process scale-up

Although a highly directional laser-induced plasma plume is advantageous for small

area deposition, it poses a problem when process scale-up is attempted. Given the

limitation on the maximum beam size with a sufficient fluence that can be generated from a

laser (few mm2), the area for film deposition is limited to only a few centimeters. In

contrast, the maximum target diameter in sputter deposition is on the order of several cm,

and therefore capable of producing a comparable size film. The deposition rate of

sputtering, on the other hand, is several orders of magnitude lower than that of laser

evaporation. It has been demonstrated that deposition rates as high as 150A/sec are

possible with PLD, using a high repetition rate.54 Although beam scanning, in

combination with substrate scanning and rotation, has been demonstrated to produce

relatively large deposition areas (2 inches in diameter), a great deal of work is still needed

to produce the desired film and compositional uniformities due to the inherent super-cosine

distribution of the plume.

1.2.5.4 Substrate-film interface reactions

As the result of high deposition temperatures, as high as 800 oC, substrate-film

interface reactions can become significant. For some applications, such as one that

involves silicon substrate, this problem becomes more pronounced. In addition, there is

the inherent difficulty of maintaining a uniform temperature as substrate radiation losses

become important at high temperature. The laser deposition technique, having a highly

energetic plasma plume, could potentially be adopted to produce films at much lower

substrate temperatures.








1.3 Overview of This Work

This dissertation addresses several of the problems associated with the PLD

technique. Though most of the studies reported in the literature address issues related to

processes occurring at the substrate and the properties of the deposited film, this work

focuses on interactions between the laser beam, the target, and plasma plume. It is thought

that a better understanding of the processes that occur at the target will lead to solutions to

some of the important problems. In general, the research reported in this dissertation is

divided into three main topics, each of which is summarized below.

1.3.1 Target Surface Morphology

The fundamental mechanisms of the development YBCO target surface features

were investigated. The effect of fluence, repetition rate, pulse number, target density,

surface roughness, and background pressure were studied. The formation of surface

features on the YBCO targets was compared to other ceramic, metal, and polymer targets.

A detailed fluence-dependent optical analysis was used to investigate the mechanism for the

surface evolution. The effect of bulk target temperature on surface morphology

development was also investigated. The physical and chemical properties of the target,

such as thermal conductivity, heat capacity, and phase transition temperatures, were

measured and related to the development of target surface morphology.

1.3.2 Particulate Reduction


The fundamental mechanism of the formation of particulates during the evaporation

of YBCO superconductors was investigated. The target temperature was found to have a

strong influence on the production of particulates. The particulate size was estimated using

a model based on the ejection of liquid droplets from the molten layer. The patterns







generated by the plume redeposited on the surface during the evaporation at atmospheric

pressure were used to provide the extent of shock wave formation.

1.3.3 Target and Film Stoichiometry

The composition of the plasma plume was examined using a mass spectrometer at

an intermediate pressure typical to most film deposition processes. The composition of the

plasma phase, including neutrals and ions, was studied as a function of pulse number,

fluence, and pressure. The ideal plasma composition under equilibrium conditions was

determined using complex chemical equilibrium analysis of the plasma phase based on a

stoichiometric algorithm. The effects of the temperature and pressure of the plasma on the

composition were investigated with this thermodynamic model and related to the

experimental results.












CHAPTER 2
EQUILIBRIUM ANALYSIS OF THE YBCO SYSTEM

2.1 Introduction

One of the advantages of pulsed laser deposition (PLD) is that the process is highly

nonequilibrium, so a multielement compound can be deposited with the same stoichiometry

as the target. This process is ideal for a compound that melts incongruently, since the

intermediate phases are not formed. In essence, since the evaporation process is highly

energetic, the solid phase is completely transformed at the target into a vapor phase with the

same composition as the target and transported to the substrate where it often condenses as

a stoichiometric compound. Because of the instantaneous absorption of the high-energy

beam, the material is superheated and becomes a one-phase plasma system.

The degree of equilibrium in a laser-induced evaporation process, however,

depends on the type of laser used in the evaporation, as well as the applied energy density.

For example, the evaporative mechanism using a pulsed UV laser with high peak energy at

a very short duration is believed to be highly nonequilibrium.4 In this case, the rapid

heating in a confined to a small volume, due to the shallow optical penetration depth ofUV

light, allowing the irradiated target temperature to increase rapidly causing evaporation. On

the other hand, for a continuous infrared (IR) laser with lower peak energy, the process is

closer to equilibrium or, in principle, is similar to a thermal evaporation process. With IR

heating, the heated target volume is large due to the deep optical penetration depth, which

can be several orders of magnitude greater than that of the UV. As the result, the

temperature increase is sluggish. In the near equilibrium process, the composition of the

solid and the vapor phases change with time.







It has been experimentally demonstrated that with PLD stoichiometry can be

preserved with up to six elements under certain conditions.3 This is only possible at a very

high fluence and at a deposition temperature low enough to yield high sticking coefficients.

At lower fluence, the stoichiometry is not preserved. For the evaporation of a YBCO

material, a lower fluence is known to produce a yttrium-deficient film. This behavior

suggests that the process is not totally physical even under the conditions of energetic UV

vaporization, and may be connected to the target surface temperature or the temperature of

the plasma.

The PLD process can be broken into two connected processes, one occurring

during irradiation at the target vapor interface and one occurring after the pulse, primarily

with vapor phases. The first process is a high-temperature, high-pressure vapor system.

This superheated vapor may be in partial chemical equilibrium, since the laser pulse

duration is very short (i.e., in nanoseconds), although the temperature is high, a condition

that favors rapid chemical reaction. Such rapid thermal increase causes superheating such

that there is no distinction between superheated solid and highly condensed gas.8 A simple

calculation of the resulting dense plasma, assuming all the photon energy is absorbed to

cause solid-vapor phase transformation, shows that the temperature could easily reach a

few thousand degrees, with pressure as high as thousands of atmospheres. This

superheated gas is more appropriately called a plasma, since a fraction of them is ionized.

At a temperature of 10,000 K every gas is ionized.55

In the second process, the plasma expands during or following the laser pulse.

This process may be closer to chemical equilibrium than the first, depending on the number

of collisions. As the hot, expanding plasma leaves the target surface, it undergoes adiabatic

expansion. A simple calculation involving a plasma that expands adiabatically shows that,

within the expansion distance, typically approximately 3 cm between the target and the

surface, the plasma could reach ambient pressure and temperature. Because of the long

transit time (as the plasma expands prior to being deposited on a substrate), which can







reach several microseconds, and the collisions with the background gas, it is likely that

chemical equilibrium can be reached.

2.1.1 Chemical Equilibrium Analysis

Chemical equilibrium analysis can be used to understand the characteristics of the

plasma generated from pulsed laser evaporation as it forms and expands. This analysis can

predict the composition of the vapor phase as a function of temperature and pressure.

Because thermodynamic constraints are insensitive to a certain dynamical variables, a

thermodynamic approach cannot be expected to give a detailed answer, however, it helps in

representing the ideal condition. A detailed discussion of thermodynamic of unstable

plasma is available.56

In addition to simple analysis, equilibrium analysis can be used to bound the

problem, namely, the extreme high-temperature, high-pressure process of the plasma

formation, and the low-temperature, low-pressure process of the plasma expansion.

Moreover, the direction in which the deposition characteristics may turn out, for example,

as the target composition is changed or as the temperature or pressure is varied, is often

predicted by an equilibrium analysis, making this analysis an easy probe tool. The

equilibrium composition can be used to predict optimum plasma temperature and pressure

for the deposition, determine physical and transport properties used in process analysis,

determine the effect of oxygen pressure and the inert gas, and to predict the species

available for the deposition process.

The author is not aware of any previous attempts to apply equilibrium analysis to

PLD. Baldwin examined laser evaporation of a binary compound and related the results to

an equilibrium analysis.57 Using a Q-switched Nd:glass laser, it was found that the

composition of deposits collected from the evaporation corresponds to the equilibrium

composition of liquid in equilibrium with the solid. A more complex thermodynamic

equilibrium simulation has been performed in the YBCO-C1,I system as applied to chemical







vapor deposition (CVD) of YBCO.58 However, the chemical species involved were

different, since metallo-organic precursors were used, and the typical temperature range for

CVD is much lower than that of the target-vapor interface during laser evaporation.

The results of the analysis presented in this chapter will be used to help interpret the

result presented in Chapter 8, where the plasma composition predicted by the computer

simulation and the experimental measurements will be compared. The equilibrium

composition is determined from the relative free energy of the possible components in the

system. The key assumptions used in the equilibrium calculations will be discussed below.

2.1.2 Fundamental Assumptions

The basic assumptions necessary to calculate plasma thermodynamic equilibrium

are local thermodynamic equilibria, electrical charge neutrality, and ideal gas. This

calculation assumed that the evaporation process is strictly thermal; therefore, the effect of

photochemistry, both photon absorption and emission, is ignored. It is further assumed

that the plasma has a uniform pressure and density. This section describes a selection of

fundamental concepts pertinent to the assumptions made above. Local equilibria and

electrical charge neutrality, the concept of temperature and its significance in the partial or

near equilibrium condition, and the significance of plasma electronic excitation in the laser-

induced plasma will be briefly discussed.

2.1.2.1 System temperature

In thermodynamic equilibrium, the distribution of translational, rotational,

vibrational, and electronic energies, as well as the distribution of energy in the spectral

radiation depend on a single parameter-the system temperature. Although the rate and the

type of energy transfer from one form to another may vary, or through which the

dissociations are achieved, the final energy distribution is strictly determined by the

temperature. The equilibrium distribution does not depend on the interactions between the

individual species, it depends only on the temperature and molecular properties of each







species.59 Once the system temperature is determined, the degree of thermal equilibrium

can be determined.

A system achieves thermodynamic equilibrium if sufficient time is allowed for it to

reach a uniform temperature in a heat bath. In the case of the vapor species produced by

laser evaporation, each vapor species, in principle, is surrounded by the plasma plume,

which acts as the heat bath. At very high temperatures, chemical reactions can proceed

very rapidly, so that the time required to approach chemical equilibrium is short, even

compared to the dynamics of the plasma transport. As a result, chemical equilibrium may

be achieved at the target surface, the highest temperature region in the process.

If the heat transfer rate in the plasma is slow, the reaction temperature is different

from the process temperature and a thermal equilibrium is not met. Where the reaction

temperature conforms to the process temperature, all species are in a state of thermal

ionization, and the plume is said to be in a state of Saha equilibrium.59

2.1.2.2 Local equilibria

Local thermodynamic equilibria in laser-induced vapor is analogous to that in metal

vapor in flame.59 Ideally, a system having thermodynamic equilibria must be adiabatic

(i.e., absence of material and energy exchange with the surroundings), while at the same

time being allowed to relax toward the state of equilibrium. Real flames are not really

adiabatic. In addition to the energy loss due to radiation and convection, fresh supplies of

combustion material are continuously added. As a result, the system consists of

temperature and concentration gradients, and therefore it is not possible for a single

temperature to represent the system. If, however, the rate of energy loss is slow compared

to the rate of energy is partition to the various degree of freedom, a concept of local

thermodynamic equilibrium may be introduced.59 The system of interest in PLD can be

assumed to be in local equilibria, since the pulse duration is short, and the transient nature

of the evaporation may prevent true equilibria from being achieved. At high temperature,







however, an equilibrium condition is more likely because the electron-ion thermalization

time varies inversely to the cubic of the temperature.60

2.1.2.3 Charge neutrality

A plasma can be defined as a collection of free-charged and neutral particles such

that the net uncompensated charge is small compared with the charge of either sign, i.e.,

plasma is electrically neutral.61 A system may be classified as plasma if the electron

density exceeds 108 cm-3.59 In a charge-neutral system, the density of electrons and the

charge density of all negative ions balance the charge density of all positive ions.

It is also possible that electrical neutrality can be achieved locally, especially in the

case where the Debye shielding distance is small, lower than 10-2 cm at an electron

concentration above 108 cm-3.55 The positive and negative charges are constrained to

move together, otherwise the electrostatic forces caused by the charge separation prevent

the movement of the charged ions.

At low temperature, the composition of ions can generally be neglected, especially

at temperatures much lower than 5000 K.55 However, at higher temperatures, especially

when oxygen is involved, ionization becomes important, and a significant concentration of

free electrons exists as the result of the equilibrium of

MxOy <--> x M++ y O++ (x+y) e- (2-1)

where M and O represent the constituents of a neutral molecule with stoichiometric

coefficients x and y, M+ and 0+ are the ions, with e- as the electron. The equilibrium

constant, K, can be given by the Saha equation, which is expressed as a function of

ionization potential, temperature, and the statistical weight of the products and reactants. It

is important to point out that treating ionization as a chemical equilibrium with the electron

as the product, implies that the degree of ionization of a metal depends on the presence of

other ions in the plume. Determining the degree of ionization of a species requires a

calculation involving the dissociation constant of other ions and the mass balance equation

of







M M+ + 0+= Y e- (2-2)

As the result of atoms-ions equilibria, the composition of the ions varies in a complex

manner.

2.1.2.4 Electronic excitation

One factor that may affect the equilibrium calculation is the presence of other

reaction paths, such as photochemistry, which may coexist with the thermal process. The

transition between two electronic states involves the gain and loss of energy in a species

from the ground state to the excited state. If the energy difference between these transition

states is small, the transition is thermal and involves a change in vibrational and rotational

states. If the energy difference is greater, the absorption and emission of radiation is

involved. The absorption of light in the visible and ultraviolet region involves the transition

between ground singlet states and some excited states of the molecules. In this energy

range, the absorption is comparable to the atomic bond strength in a molecule, which leads

to electronic excitation, bond rearrangement, and chemical reaction. The difference

between the ground state and the first excited state of a molecule is typically 150 to 600

kJ/mole.62 The transition from a singlet state to the triplet state is quantum-mechanically

forbidden since it requires spin transition. If this transition takes place, the transition from

the triplet to singlet emits photons, known as phosphorescence and has a lifetime of 10-5

sec. The travel time from the target surface to the substrate, in comparison, is

approximately 3x10-6 sec. The transition from singlet excited state to ground state also

emits photons, known as fluorescence, and has a lifetime of 10-8 sec. Due to the long

lifetime of the triplet state, the transition from the triplet to singlet states takes place without

the radiation of light. The energy, instead, is degraded into heat. The increase in

temperature, according to the Boltzman relationship, should increase the concentration of

excited atoms. However, as a result of this ionization, the number of neutral species also

decreases. Therefore, the emissions may be decreased with a hotter flame. The absorption

of photon energy by the solid and vapor during PLD involves the change of the species







from the ground state to the excited state, followed by photon emission. This emission,

however, is small and will be ignored in the simulation. The dissociation reaction

involving metal oxides plays an important part in the emission spectra of this element.

2.1.3 Theoretical Backgrounds

The chemical equilibrium condition in this study was formulated in terms of the

chemical potential. The potential function was defined as state function; that is, the

functions between the two states is independent of the path. Depending on the appropriate

constraints, entropy is at a local maximum, whereas the Gibbs function is at a local

minimum.

There are two formulations for the equilibrium conditions: stoichiometric and

nonstoichiometric formulations.63 The stoichiometric formulation is one in which the

closed-system constraint is treated by the stoichiometric technique; for the

nonstoichiometric formulation, it is treated by the Lagrange multipliers. The formulation of

the stoichiometric matrix as well as the stoichiometric restrictions have been discussed in

detail.63 Both of these formulations work by minimizing the Gibbs free energy, G, as a

function of the temperature and pressure.

( a\ =,0
ShP (2-3)

where is the extent of reaction.

The free-energy information can be presented in several ways, regardless of the

methods to be used in determining the equilibrium composition. They are the free energy

based on the Raoult and Henry conventions, the free-energy function, the conventional

absolute entropy in combination with the enthalpy of formation, and the standard electrode

potentials. In this calculation, the free-energy based on the conventional absolute entropies

(So), together with the enthalpies of formation (AHf) was used.







The value of AGO can be determined by
AGO = AHO TASo (2-4)

where
ASo = L)uiSi (2-5)

and
AHo = IZiHiO (2-6)

and Sio is the absolute entropy of species i, and ui is the stoichiometric coefficient i.

From all the above equations, it follows that
io = AHfio TSio (2-7)
If, however, the ASo and AHo are expressed in terms of the standard heat capacity at
constant pressure, Cp(T), the standard enthalpy of formation at 298 K, AHfi,298, and the

standard absolute entropy at 298 K, Si,298, such that
(T
AH,-T=AH?,298 + Cp (T) dT
f298 (2-8)

and
/T
o o= CP O)
ST=S298+ J-dT
S 298 T (2-9)

then from 2-7, 2-8 and 2-9, the following expression can be obtained

S= AH-, 298 + T Cp(T) dT T 98 + dT (2-10)
98 98 T (2-10)

The database of all the species in this simulation consist of AHof,298, S0298, and Cp(T),

which are obtained from various sources.

2.1.4 Outline of the Computer Program

The computer program used in this calculation was originally developed at the
University of Florida.64 Some modifications were made to make the program more







adaptable with the database available in this research. A brief summary of the program

algorithms is presented below.
The computer program takes the input in the form of AHof,298, S0298, and Cp(T),

together with the initial moles of each species and the elemental composition. The

temperature and pressure for the simulation can be set arbitrarily. Distinctions are made for

the vapor, liquid, solid, condense phase, and solution.

The program begins by calculating the standard chemical potentials for all the

species and by making an initial estimate of the equilibrium compositions. The standard

chemical potentials are calculated using Equation 2-10 above. The equilibrium composition

based on the initial composition is determined after the optimum basis species and activity

coefficients are calculated. The optimum basis species matrix is tested for linear

dependency using the Gram-Schmid orthogonal algorithm, while the activity coefficients

are calculated using ideal solution theory. The extent of reactions are adjusted using

nonnegativity constraints. Next, a comparison is made between these equilibrium

constants and the equilibrium constants calculated from the Gibbs free energy. If the error

is greater than the error set for the convergence criteria, the calculations are repeated until

convergence criteria are met. The details of these programs are available.64
The programs used to generate AHOf,298, S0298, and Cp(T) are mainly multiple

linear regression techniques. Since all the references are available in tabular form, the

multiple linear regression technique is used to generate the coefficients, which can be used

with the equilibrium program.

Since this simulation involves the presence of ions, the elemental basis used are

ions and electrons. The basis species are chosen to be those that represent the highest

ionization level, which are Y2+, Ba2+, Cu2+, 02+ and e-. A neutral species is, therefore,

considered as a combination of a cation and two electrons, while a negative ion is

considered as a combination of a cation with three electrons.








2.1.5 Statement of Purpose

The purpose of this study was to determine the composition of the plasma generated

from pulsed laser evaporation during the formation and expansion stages. The chemical

equilibrium calculations were performed for conditions bounded by the high-temperature

high-pressure and the low-temperature low-pressure systems. The effect of excess oxygen

pressure and inert gas concentration on the equilibrium composition was also studied. It

should be noted that the effect of photochemistry due to the laser photon was neglected.

2.2 Thermodynamic Calculations

The thermodynamic properties used in the simulations were taken from a number of

sources. The Journal of Physical and Chemical Reference Data and IVTAN Thermo

provided most of the thermophysical properties used in this calculation. Additional

thermophysical properties searches were done through STN International network, which

further searches BEILSTEIN, CA, DETHERM, GMELIN, JANAF, NISTTHERMO,

TRCTHERMO, and DIPPR.

The following thermodynamic properties, AHOf,298, So298, and Cp(T), were

extracted from the sources. The thermodynamic data were checked for internal

consistency. If only partial information was available, an estimate was made. The species

of interest are grouped into each basic element, Y, Ba, Cu, and 0, with the corresponding

states, solid, liquid, and vapor. Each entry in the table contained one to several references.

2.2.1 Data Fitting Functions


Three types of fitting functions were used for the heat of capacity at constant

pressure. These functions were chosen because, given any set of heat capacity values with

their corresponding temperature, at least one of them could give a good fit. These

functions are







Cp(T) = AO + A1.T + A2T-2 + A3.lnT (2-11)

Cp(T) = Ao + A1.T + A2.T2 + A3.T3 (2-12)

Cp(T) = Ao + A .T + A2.T2 + A3.T-2 (2-13)
The effectiveness of these functions is illustrated in Figure 2-1 for the heat capacity fitting

of 0+. A fit using Equation 2-12 gave a correlation coefficient value of 0.98, whereas for

the other two the value was only 0.80.

Each of the initial set of heat capacity data at each temperature was fitted to these

functions, and the one with the highest correlation coefficient was chosen. The heat
capacity coefficients, AHof.298, So298, the reference source, and the fit standard deviations

are summarized in Table 2-1.65-82 All the data in this table have an accuracy better than

100 K, determine from the temperature difference between the calculated and the literature

Gibbs free energy of the same value.

Where the heat capacity function did not produce an accurate fit, the calculated

Gibbs free energy may have been too far from the literature value. If the standard
deviation was greater than 100 K, new extrapolated values of AHOf,298 and S0298, which

minimized the error, were used. In this case, the standard deviation was denoted by a
symbol "d". In the case of an ion for which only the AHOf,298 value was available, the

S0298 and Cp(T) were estimated to be the same as those of the neutral species, and the

standard deviation is denoted with a symbol "e."

To evaluate the accuracy of these coefficients and the others, simulations of the

solid-liquid-vapor phase equilibria were performed for elemental yttrium, barium, and

copper (Sections 2.2.5, 2.2.6 and 2.2.7). The phase transition temperatures were then

compared with the literature values.

2.2.2 Condensed Phases


The solid phases containing at least one of the elements in YBCO are tabulated in

Table 2-2.83 All these solid phases have been identified and formed at temperatures































El,


I') ..... ... l. _J. 1 i ...._ J! 1 ] I.__,_. ___[ ~ .I _.. .___L._ .
0 2 1 6 8 10 12 11 lb 18 2(1
(Thousands)
TEMPERATURE (K)
Cp) (exp) f Cp(Fitul) o Cp(Fita2) A Cp(Fit.ld)













Figure 2-1 Temperature dependent heat capacity of O+. Comparison of three
different fitting functions.









Table 2-1 Thermodynamic Coefficients of the Species in the YBCO Chemical Equilibria
SPECIES AHf,298 SO298 AO At A2 A3 Fits Ref. Note
(kcal/mole) (kcal/moleOK)


Ba (s)
Ba (1)
Ba (g)
Ba+ (g)
Ba2+(g)
BaO (s)
BaO (1)
BaO (g)
BaO+ (g)
BaO2 (s)
Cu (s)
Cu (1)
Cu (g)
Cu+ (g)
Cu2+ (g)
Cu- (g)
Cu2 (g)
Cu2+ (g)
CuO (s)
CuO (g)
CuO+ (g)
Cu20 (s)
Cu2O (1)
e- (g)
0 (g)
0+ (g)
02+ (g)
o- (g)
02 (g)
02+ (g)
02 (g)
03 (g)
Y (s)
Y (1)
Y (g)
Y+ (g)
Y2+ (g)
YO (g)
YO+ (g)
Y203 (s)


0.00000E+O0
1.19201E+,00
4.28204E+01
1.64545E+02
3.96860E+02
-1.31063E+02
-1.17558E+02
-2.82037E+01
1.22359E+02
-1.51672E+02
O.OOOOOE+00
2.83525E+00
8.07269E+01
2.60469E+02
7.29930Ei02
5.09134E+01
1.16055E-02
2.96030E+02
-3.73178E+01
7.49878E+01
2.55000E+02
-4.08195E+01
-2.67802E+01
O.OOOOOE+00
5.9625 1E+01
3.75130E+02
1.18726E+03
2.43015E+01
0.00000E+00
2.80208E+02
-1. 14904E+01
3.40259E+01
0.OOOOOE+00
-1.15864E-01
1.01339E+02
2.46327E+02
5.29383E+02
-1.09445E+01
1.29222E+02
-4.55598E+02


1.49390E-02
1.58604E-02
4.07090E-02
4.20870E-02
4.20870E-02
1.72331E-02
2.38051E-02
5.62764E-02
5.84763E-02
2.22605E-02
7.93018E-03
9.95218E-03
3.97626E-02
3.83845E-02
3.83845E-02
3.84108E-02
5.78003E-02
5.78003E-02
1.01851E-02
5.60789E-02
5.60789E-02
2.20851E-02
3.10753E-02
5.01650E-03
3.85965E-02
3.70538E-02
3.70538E-02
3.77319E-02
4.90548E-02
4.91129E-02
5.00564E-02
5.75435E-02
1.06251E-02
7.32637E-03
4.28895E-02
4.03216E-02
4.03216E-02
5.59027E-02
5.43967E-02
2.36923E-02


5.60108E-03
1.89038E-02
4.50495E-03
-2.63591E-02
-2.63591E-02
9.16463E-03
9.83678E-03
4.49375E-02
2.04991E-02
1.41060E-02
6.44572E-03
4.64409E-03
5.42674E-03
4.97961E-03
4.97961E-03
4.97000E-03
9.08461E-03
9.08461E-03
1.15973E-02
8.96267E-03
8.96267E-03
1.75347E-02
-5.42559E-02
4.97000E-03
4.75399E-03
5.08662E-03
5.08662E-03
5.03059E-03
7.03297E-03
7.65328E-03
7.73116E-03
8.46102E-03
5.81244E-03
1.03050E-02
7.31717E-03
9.95620E-03
9.95620E-03
9.08767E-03
7.54216E-03
2.81389E-02


-1.90805E-06
-1.03272E-05
-4.65545E-07
-1.68347E-06
-1.68347E-06
1.34519E-06
5.40625E-06
3.98870E-06
5.27892E-07
6.69592E-06
-1.49983E-06
4.15329E-06
-1.18703E-06
9.03273E-08
9.03273E-08
0.OOOOOE+00
1.28676E-08
1.28676E-08
1.77285E-06
-4.99690E-07
-4.99690E-07
-6.61984E-07
-3.83687E-06
0.OOOOOE+00
1.18743E-07
-1.88197E-07
-1.88197E-07
8.12100E-10
1.19695E-06
2.45797E-07
8.66153E-07
3.62353E-06
1.62982E-06
-1.91355E-13
-3.61055E-06
-4.21214E-06
-4.21214E-06
-3.95467E-07
1.41692E-06
3.79680E-06


-3.17626E+02
2.86804E-09
1.39974E-09
4.08322E+02
4.08322E+02
-8.30147E+01
-1.51091E-09
-5.05254E+02
-2.35541E+02
1.24130E-11
1.98729E-09
-1.65357E-09
6.59858E-10
-1.40443E-10
-1.40443E-10
O.00000E+00
4.01306E-11
4.01306E-11
-1.81827E+02
4.49139E-10
4.49139E-10
2.87208E-09
4.88611E+02
0.00000E+00
-4.49053E-12
5.19815E-11
5.19815E-11
1.79628E+01
-1.21454E-10
6.08848E- 11
-1.11548E-10
1.06165E-09
8.50022E-11
7.64557E-04
1.56677E-09
1.30167E-09
1.30167E-09
1.69535E-10
-5.13830E-10
-1.01445E-09


9.00407E-04
-9.03101E+02
-1.98165E-13
4.84064E-03
4.84064E-03
4.66540E-04
1.35875E-13
-5.76722E-03
-1.57789E-03
-6.35853E-15
-2.53983E+01
2.06336E-13
-6.48923E-14
4.02391E-14
4.02391E-14
0.00000E+00
-3.27938E+01
-3.27938E+01
3.72076E-06
-4.22352E-14
-4.22352E-14
-2.35673E+02
1.14032E-02
0.00000E+00
4.63720E+01
-1.93666E-15
-1.93666E-15
-7.51902E-06
3.04348E-15
-3.72326E-15
3.22342E-15
-2.34939E-13
3.13799E+00
7.48732E-10
-1.42223E-13
-1.16956E-13
-1.16956E-13
-1.37948E+02
8.20905E-14
-4.11308E+02


Note: a) Std. dev. 10 OK; b) Std. dev. 50 K; c) Std. dev. 100 OK; d) AHf,298 and SO298 extrapolated; e) Estimated


2-11 65-68 a
2-13 65.66 C
2-12 65.66 a
2-11 65,69 b
2-11 65,70 e
2-11 65.71 b
2-12 65.72 d
2-11 65.73,80 C
2-11 73,74,81 b
2-12 72 d
2-13 65,67,75 b
2-12 65.67 b
2-12 65.73 b
2-12 65.73 a
2-12 73.76 e
2-11 65 a
2-13 65.74 a
2-13 65.74 e
2-11 65.67 a
2-12 65.73 b
2-12 73 e
2-13 65.67 b
2-11 65,67 b
2-11 65.82 a
2-13 65.72.82 d
2-12 65.82 b
2-12 77.82 e
2-11 65.82 b
2-12 65,78.82 a
2-12 65.82 b
2-12 65.82 b
2-12 65.67.82 d
2-13 67,68.72 a
2-11 72,79 d
2-12 72.73 a
2-12 73.74 C
2-12 73.74 e
2-13 73.74,80 b
2-12 73,81 b
2-13 72 a











Table 2-2 Solid Phases of Y-Ba-Cu-O System


Chemical Name

YBa2Cu307.x


YBa3Cu207-x
Y2BaCuO5

Y2BaO4

Y2Ba205

Y2Ba407
Y3Ba409

Y4Ba207

YCuO2

Y2Cu205


Y203

Y4Ba309

BaCuO2


BaCuO4

BaCu202

BaO

Ba2CuO3

Ba3CuO4

CuO


Y:Ba:Cu

123


132

211

210

220

240

340

420

102

202


200

430

011


011

012

010

021

031

001


Physical Properties

black, opaque, superconducting, orthorombic,
stable to 500 C


green, orthorombic, stable at high temperature

dissociates at 1400 OC

occurs at 900 1000 oC

occurs at 900 1000 oC

melts congruently at 2160 C




orthorhombic, melt incongruently at 1100 to
1150 OC

white, raw material, melt at 2410 OC


opaque, cubic, melt congruently between 1000
to 1010 OC


not stable at 1 atm

white, raw material, melt at 1918 OC

decompose above 850 OC

decompose above 850 OC

black, raw material, melt at 1326 OC







between 950 and 1010 OC, the temperature of interest for the formation of the 123 phase.

The phase diagram generated from the experiments and from theoretical calculations is

available.84 It is important to control the stoichiometry in the formation of the 123 phase

since a small change in the composition changes the stable species that may form, such as

the 132 and 211, which normally occurs as phase impurities. A detailed analysis of the

123 phase is available.83. 85

Heating YBa2Cu307 has been shown to produce Y2BaCuO5 and liquid phase.86

The temperature required is at least 900 OC for the barium-copper-rich phase and up to

1500 oC for the barium-rich phases.84 Since the 123 phase melts incongruently, the

composition of the liquid phase is different from that of the liquid phase, making melt

processing difficult

For the pure yttrium system, the equilibrium condition between the solid, liquid,

and vapor phase were accurately predicted, indicating that the database used in the

simulation was accurate. Below the melting point, 1800 K, the chemical potential of the

solid was lower than that of the liquid, indicating that the solid phase was more stable

(Figure 2-2). On the other hand, above the melting point, the chemical potential of the

liquid was lower, indicating that the liquid phase is more stable at this condition. Similarly,

the chemical potential of the liquid phase was lower than that of the vapor below the boiling

point of 3600 K, and higher above the boiling point. The literature values are 1800 K for

the melting point and 3607 K for the boiling point72

Phase-transformation simulation of barium metal showed a melting point of 935 K

and a boiling point of 2115 K. The reference values are 1002 K and 2167 K,

respectively.72

The phase transition temperatures for a copper element were calculated. The

melting point was calculated at 1325 K, whereas the boiling point was at 2825 K. The

literature values for the melting and boiling point temperatures of the copper element are

1358 and 2800 K, respectively.72





















0-





%

0



Melting Point


w \
O



36 180000 oK






-150 -
Y (S)

.......... Y (L)

S-Y(V)
S-1200
0 1000 200 3000 400 500 60












TEMPERATURE (K)
y Boiling Point \ ,

3600 oK \ ,,


-150 Y(S) \ "

(L) \ "






-200 --1-
0 1000 2000 3000 4000 5000 6000




TEMPERATURE (K)


Gibbs free energy yttrium element as a function of temperature.


Figure 2-2







2.2.3 Gas Phase

The gas species that contain at least one of the elements in YBCO and have been

reported in the literature are presented at Table 2-3. Most of these species are atoms and

oxides, both as neutral or charged species. Since the temperature was very high, beyond

the boiling points of all the species, only the gas-phase reactions were considered in the

simulation. The system pressure was chosen to be 1 atm, since this was an intermediate

pressure that may best represent the plasma. The plasma pressure varies from tens of

atmospheres on the target surface to fractions of an atmosphere as the plasma expands.

The general equilibrium reactions in the vapor phase can be summarized using the

following equations:
MO <+ M + O AHOxn = 3-7 eV (2-14)

MO + MO+ + e- AHoxn = 5-7 eV (2-15)

M +-+ M+ + e- AHOrn = 5-8 eV (2-16)

M+ +- M2+ + e- AHOxn = 10-20 eV (2-17)

O -> 0+ + e- AHrxn = 14 eV (2-18)

O+ +-- 02+ + e- AHon = 35 eV (2-19)

where M is Y, Ba, or Cu. The heat of reaction values were determined to be 298 K. All of

the lower and upper limits on the heat of reaction values are for the Ba and Cu,

respectively, except for Equation 2-14, which correspond to Cu and Y, respectively.

2.2.4 Oxygen System

Simulation of the oxygen system, consisted of the following vapor species: O, 02,

03, 0+, 2+, O-, 02+, and 02-. The pressure was kept at 1 atm, whereas the temperature

was varied from 298 to 20,000 K (Figure 2-3). The four dominant species in this phase

equilibria, which exist with mole fractions greater than 10-2, are 02, O, 0+ and e-. At a

temperature below 3500 K, 02 is the majority species in the plasma. At temperatures








Table 2-3 List of vapor species detected in laser induced YBCO plasma


NO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25


MS=mass spectroscopy; ES=emission spectroscopy; PL-photoluminesence
spectroscopy; FS=fluoresence spectroscopy.


SPECIES
Y
Y+
y2+
y3+
YO
YO+
Y02+
Y022+
Y20
Ba
Ba+
Ba2+
BaO
BaO+
Ba20
Ba202
Cu
Cu+
Cu2+
CuO
CuO+
CuO2+
O
0+
02+


DETECTION METHOD
MS, PL, FS
PL, FS
ES
ES
MS, PL, FS
FS
MS
MS
MS
MS, PL, FS
PL, FS
ES
MS, PL
ES
MS
MS
MS, PL
PL
ES
MS, PL
ES
MS
MS, PL
PL, FS
PL, FS


Note:


















101



100



10 1



10-2




10-3



10-4


10-5




10-6


10' 1



10"8

10 -

1/I 02-



10-. ,_, .
0 5000 10000 15000

TEMPERATURE (K)


Oxygen phase equilibria at 1 atm.


20000


Figure 2-3







between 3500 and 16,000 K, O species predominates, whereas beyond 16,000 K, 0+ and

e- ions predominate. The concentration of 03 species reaches a maximum at 3000 K, then

disappears with increasing temperature. 02+ and 0- are comparable in concentration at

temperatures beyond 5000 K. While the concentration of 02' is negligible, in the ppb

range, 02+ is practically nonexistent due to its high ionization potential (35 eV).

If the plasma temperature generated from laser evaporation is indeed around 8000

K, as commonly stated in the literature, then the equilibrium plasma consists mainly of O

atoms, with decreasing 02 and increasing 0+, two to three orders of magnitude lower than

that of the O atoms. A higher concentration of ions, if detected, therefore can only be

attributed to the photochemistry process.

2.2.5 Yttrium-Oxygen System

The yttrium-oxygen system equilibrium diagram with excess oxygen is presented in

Figure 2-4. All the species in this figure have mole fractions greater than 10-2 at some

temperature. The oxygen species, 02, 0, and 0+, the dominant species shown in Figure

2-3, showed temperature-dependent concentration profiles identical to those of a pure

oxygen system. The values of entropy and heat capacity of Y2+ was assumed to be the

same as those of Y+ since only the enthalpy value was available. Both Y and YO species

exist at high concentration at temperatures below 13,000 and 9000 K respectively. It has

been reported that YO transfer is the main path for oxygen transfer to the film.87 The YO+

cation exists only in the temperature range of 3000 to 11,000 K, with a maximum at 7000

K. The Y+ exists at temperatures beyond 4000 K with a maximum at 10000 K, while the
Y2+ only exists at temperatures beyond 10000 K. In the temperature range of interest in

laser deposition, around 8000 K, the major species are Y, Y+, YO and YO+, with

minimum concentration Y2+, consistent with the composition of plasma observed using

emission spectra.88 The thermodynamic properties of yttrium-oxygen system, including

YO up to 3000 K has been studied.89 90
























































0 5000 10000 15000


TEMPERATURE (K)


Yttrium-oxygen vapor phase chemical equilibria at 1 atm.


20000


Figure 2-4







2.2.6 Barium-Oxygen System

The equilibrium composition of a barium-oxygen system with excess oxygen is

presented in Figure 2-5. The composition of the oxygen species, 02, 0, and 0+, is

identical to the one in Figure 2-3. The BaO species dominates the vapor phase until 6000

K because, compared to the other species, it has the lowest heat of formation. It had been

experimentally shown that the concentration of higher gaseous barium oxides, Ba20,

Ba202, Ba203 and Ba204 were negligible in comparison to BaO,91 therefore they were

ignored in this calculation. The concentration of Ba vapor increases with increasing

temperature beyond its boiling point temperature (2000 K) until it reaches a maximum at

5000 K. At temperatures beyond 4000 K, an increasing number of the Ba atoms are

converted to Ba+ ion. The Ba2+ ion concentration increases at 7000 K until it saturates the

plasma at 10000 K. The Ba+ ion, in comparison, reaches a maximum at 7000 K and starts

to decrease at 10000 K. At the temperature of interest in laser evaporation, around 8000 K,
Ba+ is at its maximum, while the Ba2+ is reaching a saturation point. Both the BaO and

BaO+ are rapidly disappearing at the same point.


2.2.7 Copper-Oxygen System


The equilibrium composition of a copper-oxygen system with excess oxygen is

presented at Figure 2-6. The composition of the oxygen species, 02, 0, and 0+, was

identical to the one in Figure 2-3. Beyond the boiling point temperature of 2800 K, the

composition of Cu and Cu2 vapor decreases with increasing temperature, whereas the

concentration of Cu+ increases. The entropy value for Cu+ reported in JANAF table 65

was significantly larger than others, therefore it was not used. Because only enthalpy value

was available, 76 the entropy and heat capacity for Cu2+ were assumed to be the same as

those of Cu+. The concentration of Cu-, although very low, reaches a maximum at 5500

K. The Cu2+ ion concentration becomes significant at temperatures beyond 15000 K.















101

BaO
100


1 -~Ile-o.W


10-2 / Ba 2+ /





10

10,

1 0











10,8 / 0+
I



Ba+B





10-1 I i
0 5000 10000 15000


TEMPERATURE (K)


Barium-oxygen vapor phase chemical equilibria at 1 atm.


20000


Figure 2-5















10 0


0 5000 10000 15000


TEMPERATURE (K)


Copper-oxygen vapor phase chemical equilbria at 1 atm.


20000


Figure 2-6







Because of their high heat of formation values, both CuO and CuO only occur in small

concentration and disappear at temperatures beyond 5000 K, consistent with experimental

result.92 The ionization potential for CuO was estimated to be the same of that of Cu since

none of the thermophysical property for the CuO* is not available. This estimate can be

justified since the ionization potentials of all of the oxides shown in Table 2-1 or similarly,

Equations 2-15 and 2-16, are comparable to those of the elements. This high ionization

potential value is reflected in the mole fraction of CuO+ is only in the ppm range at its

maximum. In the temperature range of interest in laser evaporation, the Cu+ ion dominates

the vapor phase. The concentrations of CuO, CuO+, Cu-, Cu2 and Cu2+ have decreased to

below one ppm. The concentration of Cu2+ is increasing but it will not become significant

until 15000 K.

2.3 Complex Reaction Equilibria at Elevated Temperature

The database used to simulate the YBCO system is presented at Table 2-1. The

initial concentration of the chemical species adds up to the concentration of the

stoichiometric YBCO. Some oxygen and inert gas were added to study the effects of the

environments. The calculation for the YBCO system, in principle, is a composite of the

calculations of the Y-O, Ba-O,and Cu-O systems presented above, since all the elements,

with the exception of oxygen, do not interact. The difference, however, is that the oxygen

concentration was kept at stoichiometric YBCO. For graphical clarity, only the significant

species that exist at mole fractions greater than 10-3 will be presented.

2.3.1 Effect of Temperature


The effect of temperature on the composition of Y, Ba, Cu and O system was

investigated. The process simulation was a solid YBCO superconductor, which transforms

to the vapor phase by maintaining the material balance. The pressure of the system was

maintained at 1 atm, whereas the temperatures were varied from 3000 to 20,000 K. The







result of the simulation for the species present at mole fractions higher than 10-3 is

presented at Figure 2-7.

In general, all the oxide species, both neutral and charged, have disappeared within

the first 6000 K. The CuO species, in particular, only exist up to 3000 K. Beyond 6000

K, an increasing number of charged elemental species dominate the vapor phase.

Correspondingly, the electron concentration also increases. Only about one-tenth of all the

species are charged at 5000 K, whereas at 16000 K, almost half of all the species are

charged.

All the neutral elements transform to ionic species at temperatures beyond 6000 K.

For the Y element, the transformation from Y to Y+ and from Y+ to Y2+ take place at 8000

and 13000 K, respectively. For the Ba element, the transformation of Ba to Ba+ and Ba+

to Ba2+ takes place at 6500 and 11500 K, respectively. And lastly, for the Cu element, the

transformation of Cu to Cu+ and Cu+ to Cu2+ takes place at 9500 and a temperature higher

than 20,000 K, respectively. The concentration of the ionic species increases and reaches a

saturation value at high temperatures.

For the deposition of thin film, it is desirable to have the temperature below 6000

K,where the concentration of oxides is still abundant. It can be concluded that in most

deposition processes, where only ionic species are found, the temperature of the plasma is

greater than 7500 K.

2.3.2 Effect of Pressure


The pressure dependence of the copper boiling points was studied from the range of

1 Torr to 10 atm to test the accuracy of the program and the database (Table 2-4 69). The

result was found to be consistent with the literature, even under very low and very high

pressures. The deviation at the very high pressure is caused by the system nonideality.

The effect of system pressure on the equilibrium composition of a YBCO system

was investigated. The results for the pressure of 0.2, 0.5, 1, 10 and 50 atm are presented


























































.001 L-
2~S00


5000 7500 10000 12500 15000 17500 20000


TEMPERATURE (K)






Figure 2-7 YBCO vapor phase equilibria at temperatures from 2500 to 20000
OK and pressure 1 atm.

















Table 2-4 Pressure dependence on the boiling point temperature of the copper element

Pressure Tb Experimental Tb Calculated
(torr) (OK) (OK)
1 -- 1685

10 2143 2130

100 2463 2448

400 2713 2712

760 2873 2837

7600 3773 3450







at Figures 2-8 to 2-12. Note that at pressure greater than 10 atm, the computational

accuracy decreases as the system becomes less ideal. As predicted from the Le Chatelier

principle, as the pressure decreases, the concentration of the large molecule increases,

while the concentration of atoms decreases.

The mole fraction of electron, indicative of the total charges, decreases as the

pressure increases because the formation of neutral atom is preferred. The electron mole

fractions for a plasma at 8000 K with pressures up to 50 atm are presented in Figure 2-13.

The corresponding electron density, calculated using ideal gas law, varies from 1016 to

1018 cm-3 at temperatures between 5000 and 10000 K (Figure 2-14). The ion

concentration reached a limiting value beyond 8000 K.


2.3.3 Effect of Oxygen Pressure

The effect of oxygen pressure on the equilibrium composition of a YBCO system

was investigated. The addition of 100% excess oxygen increases the concentration of

oxides and ions, and decreases the concentration of the elements. This result is presented

in Table 2-5. Note that the underlined values indicate the reduction in concentration from

the stoichiometric composition.

In the formation of the oxides, the reaction reverse of eq. 2-15, for which M and O

are the reactant and MO is the product, the extent of reaction to form MO depends on the

product of the equilibrium constant, K, and the initial concentration of both reactants;

therefore, with the increase in the initial oxygen concentration, the formation of the oxide

increases. Accordingly, the concentration of the oxide ions increases with the increase in

oxide concentrations. As an illustration, at 8000 K, the presence of 100% excess oxygen

increases the concentration of the ion oxides by an average of 50%. The increase of the

neutral oxides, however, is only about 10 %, due to further conversions to ionic oxides.

In Equations 2-16 and 2-17, where the right-hand side of the equation is the

products, cations, and electrons, the increase in the extent of reaction can be expressed in
















1 -




















z
0
.1 -
U.




















b000
OFigure 2-8
/













.01 -










Figure 2-8


6000 7000 8000 9000 10000


TEMPERATURE (K)






YBCO vapor phase equilibria at temperatures from 5000 to 10000
oK and pressure 0.2 atm.






































z
O


..1
O
LI
S1
0


















.01
5000










Figure 2-9


6000 7000 8000 9000



TEMPERATURE(K)


10000


YBCO vapor phase equilibria at temperatures from 5000 to 10000
OK and pressure 0.5 atm

























































6000 7000 8000 9000


TEMPERATURE (K)


Figure 2-10


YBCO vapor phase equilibria at temperatures from 5000 to 10000
oK and pressure 1 atm


.01 L
5000


10000









































z

c1 N
cr
IL
w
-j











0








.01 L;--
bUOOO


6000 7000 8000 9000


10000


TEMPERATURE (K)








Figure 2-11 YBCO vapor phase equilibria at temperatures from 5000 to 10000

oK and pressure 10 atm.
































































6000 7000 8000 9000


TEMPERATURE (K)


Figure 2-12


YBCO vapor phase equilibria at temperatures from 5000 to 10000
oK and pressure 50 atm.


z
O


u.
U
-j
O
2


.01 L-
o00o


10000






























































.1 1 10


PRESSURE (atm)


Figure 2-13


Mole fraction and number density of electron in YBCO plasma at a
temperature of 8000 K and pressures up to 50 atm.


-1019













10 18


-1017













--1016
100












1019


+- -


10,000 K


8,000 K


5,000 K


PRESSURE (atm)


Figure 2-14


Electron density of YBCO plasma at temperatures of 5000, 8000
and 10000 K at pressures up to 50 atm.


S1018

.o
z
z
0
I-
tU


1017


101 6L
.1


~






Table 2-5 Equilibrium composition of YBCO vapor phase at 8000 K and
1 atm with various initial conditions


SPECIES

Y

Y+
y2+

YO
YO0

Ba

Ba+

Ba2+

BaO

BaO+

Cu

Cu+

Cu2+

CuO

CuO+

0

0+

02

02+
e-

inert


Stoichiometric


Initial Comp.
(moles)

0.10000E-01

0.10000E-01

0.10000E-01

0.10000E-01

0.10000E-01

0.20000E-01

0.20000E-01

0.20000E-01

0.20000E-01

0.20000E-01

0.30000E-01

0.30000E-01

0.30000E-01

0.30000E-01

0.30000E-01

0.40000E-01a

0.40000E-01a

0.40000E-01a

0.40000E-01a

0.32000E+00

0.00000E+00b


With excess 02 With inert gas
(moles) (moles)


(moles)

0.15966E-01

0.33834E-01

0.14800E-04

0.68193E-04

0.11680E-03

0.86638E-02

0.90273E-01

0.96078E-03

0.14605E-04

0.87674E-04

0.13349E+00

0.16489E-01

0.58327E-10

0.16374E-04

0.46778E-05

0.35962E+00

0.89915E-05

0.31323E-04

0.13561E-07

0.14277E+00

0.00000E+00


0.12815E-01

0.36913E-01

0.21948E-04

0.74899E-04

0.17438E-03

0.64981E-02

0.92033E-01

0.13314E-02

0.14990E-04

0.12231E-03

0.12841E+00

0.21560E-01

0.10366E-09

0.21553E-04

0.83695E-05

0.71939E+00

0.24449E-04

0.85742E-04

0.50459E-07

0.15354E+00

0.00000E+00


Note: a
b
c


With 100 % excess oxygen, this initial value is 1.0000E-00 moles
With 50% inert gas by volume, this initial value is 6.6000E-01 moles
Underlined values indicate decrease in concentration from normal


0.11133E-01

0.38741E-01

0.27829E-04

0.25789E-04

0.72538E-04

0.54290E-02

0.92894E-01

0.16235E-02

0.49639E-05

0.48931E-04

0.12469E+00

0.25293E-01

0.14697E-09

0.82955E-05

0.38917E-05

0.35979E+00

0.14772E-04

0.16996E-04

0.12084E-07

0.16037E+00

0.66000E+00







terms of diluent effect.93 The addition of oxygen, which is considered an inert gas in

Equations 2-16 and 2-17, increases the total volume of the system. Since the extent of

reaction is expressed in terms of the product of the equilibrium constant and the total

volume, additional oxygen favors the formation of the products. As an illustration, at 8000

K, the concentrations of M+ and M2+ increase by approximately 10 and 40% respectively.

2.3.4 Effect of Inert Gas

The effect of inert gas on the equilibrium composition of a YBCO system was

investigated. The equilibrium composition is also presented at Table 2-5. The underlined

values indicate the reduction in composition compared to the stoichiometric composition.

This result, which is similar to that for the system with excess oxygen, causes the increase

in the ion concentrations by the dilution effect93 The addition of inert species increases the

total volume of the system and increases the extent of reaction proportional to the square

root of the total volume. In this process, the reactants are the neutral elements and oxides.

The products, are all the ionic species. With the additional inert gas, which constitutes half

of the total moles, the concentrations of the neutral oxides and ionic oxides at 8000 K

decreased by about 40 and 20%, respectively, whereas for the charged elements, both the

M+ and M2+, the concentrations increased by 20 and 40%, respectively.


2.3.5 Analysis of PLD Plasma


The equilibrium condition for oxides, elements and ions in stoichiometric YBCO

are presented in Figures 2-15. In general, at high pressure and low temperature, the

formation of oxides is preferred, except for copper-oxygen where only a small

concentration of CuO was formed at pressures greater than 100 atm. At high temperature

and low pressure, the formation of ions is preferred. Because of the low ionization

potential, Ba2+ can be observed at high temperature and low pressure, whereas other ions

only exist as single charged species.


























100








U

c 10

w








1










.1 -
5000


TEMPERATURE (K)


Figure 2-15


Composition of YBCO plasma at temperatures between 5000 and
10000 K and pressures up to 50 atm. Note the formation of oxides
at low temperature and high pressure and the formation of ions at
low pressure and high temperature.


6000 7000 8000 9000


10000







In the adiabatic expansion of plasma, from high-temperature high-pressure to low-

temperature low-pressure system, it is likely that initially the plasma consist of atomic

charged species and some oxides. As the plasma expands, depending on the heat capacity

ratio, both the temperature and the pressure drops with the temperature drops faster than the

pressure, favoring the formation of oxides.

2.4 Conclusion

Chemical equilibrium analysis has been shown to be useful in predicting the

composition of the gas phase during the irradiation of the YBCO target. The calculations

can predict the composition of the vapor phase, consistent with the composition as

predicted from the plasma diagnostic. The analysis shows that the formation of oxide

species is limited to the first 6000 K. The absence of CuO species is caused by the

relatively high heat of formation. Beyond 6000 K, the vapor phase is dominated by

elements, both neutral and charged species. The change in pressure affects the equilibrium

composition, consistent with the Le Chatelier principle, which increases the transition

temperature with the increase in pressure. The addition of oxygen increases the formation

of both the oxides and the charged species, while the addition of an inert gas only increases

the production of the charged elements, and reduces both the neutral elements and the

oxides. If during the laser deposition, high concentrations of oxides are desired in order to

preserve the target film oxygen stoichiometry, it is desirable to perform the deposition at

high oxygen pressure.

In the treatment of a high-temperature high-pressure plasma which expands

adiabatically, it can be expected that initially, the plasma contains a high concentration of

neutral atoms and oxides since the concentration of neutral atoms and oxides is favored at

high pressure. As the plasma expands, both the temperature and pressure drops, favoring

the formation of ions, until the temperature drops significantly that oxide and neutral

species are again formed.












CHAPTER 3
SELECTED THERMOCHEMICAL AND THERMOPHYSICAL PROPERTIES
STUDIES


3.1 Introduction

To describe the state of an initially solid superconductor during laser evaporation,

its thermochemical and physical properties must be accurately known. Various

characterization techniques have been used to determine the physical as well as the

superconducting properties of YBCO. Most characterization, however, has been limited to

the physical properties at cryogenic temperatures, typically near the superconducting

transition temperature. In previous modelling work, the available data at cryogenic

temperatures were extrapolated to establish the physical properties at the high temperatures

required for the heat-transfer modeling of bulk superconductor and thin film preparation.

To minimize the error associated with the extrapolation of physical properties from the

cryogenic range to the melting temperature, a series of measurements at high temperature

were performed to determine the heat capacity and thermal conductivity.

This work focuses on determining these physical properties of YBCO

superconductor at temperatures close to the melting point. The measurements were

performed using Differential Thermal Analysis (DTA), Thermal Gravimetric Analysis

(TGA), Differential Scanning Calorimetry (DSC), and the Thermal Conductivity Laser

Flash Method.








3.2 Sample preparation

Samples were prepared from a ceramic powder obtained from Rhone-Poulenc. The

powder was grounded in a ball mill using 7 mm diameter zirconia beads in Freon until the

particle size was about a micron. Since the rate of sintering is inversely proportional to the

cube of the particle diameter, it is desired to have the smallest particle diameter possible.

The powder was initially pressed in a 3/4 inch diameter die at 2500 psi to produce its initial

form. The loose pellet was then isostatically pressed at 50,000 psi. This process

decreased the pellet dimensions by about 10%. Next, the pellet was sintered in dry

oxygen. A series of desicator, containing Ascarite (NaOH fine particle in silica) and

anhydrous MgC104 were used to remove CO2 and moisture from the flowing oxygen. The

pellet was then placed in a gold lined ceramic crucible to prevent a reaction with the

container materials. During the sintering process, the oven temperature was increased at a

rate of 10 oC/min to 450 oC, where it was held for 1 hr to evaporate possible organic

contaminants. The temperature was then increased to 960 OC and held there for 2 hours.

After sintering, the sample temperature was lowered at a rate of 10 oC/min to room

temperature.94

After the sintering, the pellet dimensions were further reduced by 15 %. The

sintered pellet was black and had a density of 6.0 g/cm3 or 96% of the theoretical density.

Low-density targets, 65% of the theoretical density, were made using the same procedure,

except that the powder was uniaxially pressed at 5000 psi instead of isostatically pressed at

50,000 psi. XRD analysis of this sample showed that the target was pure YBCO with the

correct oxygen stoichiometry.








3.3 Differential Thermal Analysis

3.3.1 Introduction

Differential thermal analysis (DTA) measures the heat absorbed or released by a

chemical system by measuring the temperature difference between the sample and a known

inert standard as the temperatures of both are increased at a constant rate.95 A minima in

the differential thermogram, a plot of temperature difference versus temperature, indicates

that the sample is colder than the reference sample as a consequence of an endothermic

process; a maxima indicates an exothermic process. Using a known sample as a standard,

the magnitude of the heat effect may also be determined from the peak area. It was

intended to use this technique to determine the number of reactions or phase

transformations that occur during the heating of a YBCO sample and the magnitude of the

heat of fusion.


3.3.2 Experimental

The DTA scan for the YBCO sample was performed using a DuPont 9900 General

V2.2 differential thermal analyzer. The temperature scan was from 25 to 1200 oC at the

rate of 20 OC/min. During measurement, the alumina reference crucible was left empty,

while the alumina sample crucible was filled with the sample of interest. The equipment

was initially calibrated using a gold powder sample. The superconductor powder sample

was obtained from a ground, sintered pellet. The weight of the samples used was

approximately 40 mg. The temperature was monitored during the heating and cooling

cycles.

The heat of fusion can be calculated from the area under the peak of the differential

thermogram. The peak area, A, is a function of the sample mass, m, thermal conductivity,

k, and the enthalpy of fusion, AHf, by the following relationship95







A = -(G/k)m AH (3-1)

where G is the geometric constant. The value of G was determined from measurements

with the gold sample. The heat of fusion for gold is 15.3 cal/g at 1336 K.69 The thermal

conductivities of the Au and YBCO at 400 OC are 30069 and 2.4 W/m.K, respectively.

3.3.3 Result and Discussion

Figure 3-1 shows a typical scan for YBCO powder in the temperature range of 25

to 1600 oC performed in air. A typical DTA scan available in the literature for YBCO

sample was from 25 to 1250 oC.96 In the low temperature range, the scan shows a broad

and shallow minimum, possibly due to volatilization of residual water or organic. A

broad maximum is also observed near 500 OC and is possibly related to oxygen uptake by

the sample. Three sharper peaks are observed at 950, 1040 and 1260 oC. The high

temperature region near these first two peaks is shown in greater detail in Figure 3-2 for

both the powder and pellet samples. These thermal arrests are seen in both the pellet and

powder samples indicating that the pellet formation process is not responsible for the

occurrence of these arrests. The heat effect in the pellet is somewhat larger due to the larger

mass of the pellet, though the thermal conductivity is higher. For comparison, the scan for

the gold sample is also shown in Figure 3-2.

SEM of the YBCO sample following the analysis showed that the sample had

melted and resolidified (Figure 3-3). A glassy surface and solid that had not yet melted

were observed. From the phase diagram it was expected that YBCO superconductor melt

incongruently at 1277 K84; therefore, it was possible that only a small part of the solid

underwent a phase transformation. The cooling curve showed an exothermic reaction at

1075 K.

Based on the calibration using the gold sample, the heat of fusion of YBCO was

measured to be 0.24 cal/g or 160 cal/mole. Although the area under the curve for the

YBCO sample is twice that of the gold (Figure 3-2), the thermal conductivity of the gold is

































t-
0
c

-2





a.
o -3
I--


0 200 400 600 800 1000 1200


1400 1600


Temperature (C)





Figure 3-1 Differential Thermal Analysis of YBCO powder. Heating rate: 20
oC/min. Sample size: 39.2 mg. Temperature range: 30 to 1600 OC.
Atmosphere: air.















































800 900 1000 1100


1200


Temperature (C)


Differential Thermal Analysis YBCO pellet (wt. 99.9 mg), YBCO
powder (wt. 39.2 mg) and gold powder (wt. 86.9 mg) at
temperatures around the melting point. Heating rate: 20 OC/min.
Atmosphere: air.


Figure 3-2

















































Figure 3-3 Resolidified sample of YBCO powder after undergoing DTA
analysis to 1100 OC. Magnification (a) 30x; (b) 600x.