Citation
Nucleation and epitaxy of conductive buffer on (001) Cu for coated high-temperature superconducting conductors

Material Information

Title:
Nucleation and epitaxy of conductive buffer on (001) Cu for coated high-temperature superconducting conductors
Creator:
Kim, Kyunghoon
Publication Date:
Language:
English
Physical Description:
xv, 145 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Annealing ( jstor )
Electrical resistivity ( jstor )
High temperature ( jstor )
Hypertension ( jstor )
Oxides ( jstor )
Oxygen ( jstor )
Single crystals ( jstor )
Tin ( jstor )
X ray diffraction ( jstor )
X ray film ( jstor )
Dissertations, Academic -- Materials Science and Engineering -- UF
Materials Science and Engineering thesis, Ph. D
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2005.
Bibliography:
Includes bibliographical references.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Kyunghoon Kim.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
028077802 ( ALEPH )
880637329 ( OCLC )

Downloads

This item has the following downloads:


Full Text











NUCLEATION AND EPITAXY OF CONDUCTIVE BUFFERS ON (001) Cu FOR
COATED HIGH-TEMPERATURE SUPERCONDUCTING CONDUCTORS














By

KYUNGHOON KIM


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


2005
































Copyright 2005

by

Kyunghoon Kim































To my family for their love and encouragement














ACKNOWLEDGMENTS

Above all, I would like to express my sincere appreciation to my advisor, Professor David P. Norton, for his excellent guidance and invaluable help. I cannot forget his warm support and the five years that I worked for Dr. Norton was the precious time in my life.

My appreciation also goes to my committee members, Professor Cammy R.

Abernathy, Professor Rajiv Singh, Professor Wolfgang Sigmund, and Professor Andrew Rinzler.

Since I started collaboration research in Oak Ridge National Laboratory, I received a lot of support from the staff members and post doctors. I would like to give my deep appreciation to Dr. David K. Christen, who is my supervisor and a group leader of the superconductivity group. Dr. Christen gave made me feel comfortable staying in ORNL as a graduate student. I wish to give my special thanks to the members of Condensed Matter Sciences Division, Metals and Ceramics Division, and Chemical Science Division: Dr. Claudia Cantoni, who helped me use the PLD system and shared a lot of time to discuss technical issues; Dr. Tolga Aytug, who helped me use the sputtering system and gave me great advice; Dr. Albert A. Gapud, who helped me use the transport measuring system; Professor James R. Thompson, who helped me use the SQUID system and gave me lots of technical information; Yifei Zhang, who helped me use the SQUID system; Dr. M. Paranthaman, who gave me valuable help and encouraged me; Dr. Amit Goyal, who supported metal substrates; Patrick M. Martin, who helped me use electrical measuring instruments and annealing system; Dr. Ho Nyung Lee, who helped me get a








lot of physical results and shared precious time to discuss technical issues; Dr. Sukill Kang, who helped me use various instruments and gave me warm support; and Dr. Daeho Kim and Dr. Mina Yoon who taught me lots of fundamental physics. I also want to thank all the members of Dr. Norton's group.

I would like to acknowledge the enormous help of my sister-in-law, Youri, and her family. They gave me a lot of joy and happiness. I also express my deep appreciation to my elder sister, Hyeonjoo, who always encouraged me and prayed for my family. I would like to give my special thanks to my parents-in-law for their great affection and concern. Especially, it is my great honor to dedicate this work to my father-in-law who passed away last February. I am also deeply grateful to my parents, with all my respect, for their endless love, patience and prayers for me and my family.

Finally, my great gratitude goes to my lovely wife, Youmi Kang, and my son, Youngwook Kim. I sincerely appreciate my wife's encouragement, support and love. Without her love and support, I could not do all of this.















TABLE OF CONTENTS

pnge

A C KN O W LED G M EN TS ............................................................................................ iv

L IS T O F T A B L E S ........................................................................................................... v iii

L IST O F F IG U R E S ...................................................................................................... . . ix

A B S T R A C T ..................................................................................................................... x iv

CHAPTERS

1 IN T R O D U C T IO N ........................................................................................................ 1

2 LITERA TU RE REV IEW ............................................................................................ 4

2.1 High Temperature Superconductors ................................................................. 4
2.2 Superconducting Power Applications using FITS Wire ................................... 9
2.2.1 Biaxially Textured M etal Substrates .......................................................... 11
2 .2.2 B uffer L ayers ..............................................................................................13
2.3 Review of Conducting Oxide Buffers for Cu based YBCO Coated Conductors. 16
2 .3.1 Introduction .......................................................................................... . . 16
2.3.2 Preventing Methods of Cu Oxidation .................................................... 20
2.3.3 Thin Film Techniques for Oxides Growth ............................................ 22
2.3.4 Overview of Conductive Oxides for HTS Coated Conductors ............. 25

3 GROWING EPITAXIAL Cu2Mg AS AN OXIDATION BARRIER FOR HIGH
TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS ............. 55

3.1 Introduction .................................................................................................... . . 55
3.2 E xperim ents .................................................................................................... . . 57
3.3 R esults and D iscussion ................................................................................... 57
3.4 C onclusions .................................................................................................... . . 58

4 (LaSr)TiO3 AS A CONDUCTING BUFFER LAYER FOR HIGH TEMPERATURE
SUPERCONDUCTING COATED CONDUCTORS .......................................... 66

4 .1 Introduction .................................................................................................... . . 66
4.2 Experiments ............................................... 68
4.3 R esults and D iscussion ................................................................................... 69









4 .4 C onclu sion s ........................................................................................................... 73

5 EPITAXIAL (LaSr)TiO3 AS A CONDUCTIVE BUFFER FOR Ni-W BASED HIGH
TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS ............. 90

5.1 Introduction .................................................................................................... . . 90
5.2 Experiments ............................................... 92
5.3 Results and Discussion ................................................................................... 93
5.4 C onclusions .................................................................................................... . . 96

6 EPITAXIAL (La,Sr)TiO3 AS A CONDUCTIVE BUFFER FOR Cu BASED HIGH
TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS ............ Il

6.1 Introduction .................................................................................................. . . I
6 .2 E x perim ents ........................................................................................................ 113
6.3 R esults and D iscussion ....................................................................................... 114
6.4 C onclusions .................................................................................................. . . 117

7 S U M M A R Y .................................................................................................................. 13 1

L IST O F R EFE R EN C E S ................................................................................................. 134

BIOGRAPHICAL SKETCH ........................................................................................... 145





























vii














LIST OF TABLES


Table pne

2-1. The critical temperature values in absolute temperature unit reported for YBa2(CuI.
M )30 7-6 system s ............................................................................................... 3 1

2-2. The list of buffer layer materials tested with RABiTS applications ...................... 32

2-3. The overall electrical and structural properties of perovskite oxide compounds
m entioned in this section .................................................................................... 33














LIST OF FIGURES


Figu page

2-1. (a) Schematic drawing of energy gap in superconducting material, (b) theenergy gap
dependence on temperature with theoretical value versus experimental results for
som e elem ental m etals ....................................................................................... 34

2-2. Schematic drawing of the critical space within which the superconductivity remains
(T -H - diagram ) ................................................................................................... 35

2-3. Phase diagram of type 1I superconductor with temperature and magnetic field
d ep end ency ............................................................................................................... 3 5

2-4. Schem atic draw ing of La2CuO 4 ............................................................................. 36

2-5. Schematic drawing of YBa2Cu307-i unit cell combined with oxygen deficient
perovskite structure ............................................................................................ 37

2-6. Schematic drawing of YBa2Cu306 compound which has different CuO chains with
Y B a2C u 30 7 ............................................................................................................... 3 8

2-7. Schematic drawing of (a) Bi-Sr-Cu-O, (b) Bi-Sr-Ca-Cu-O (double Cu-O sheets), and
(c) Bi-Sr-Ca-Cu-O (triple Cu-O sheets) ............................................................ 39

2-8. Conceptual drawing of rolling assisted biaxially textured substrate (RABiTS)
process based on Ni substrate. The pattern inside the circle designates the texture
alignm ent of the substrate ................................................................................... 40

2-9. The X-ray diffraction ()-scan (rocking curve) of (002) Ni which indicates the
substrate's out-of-plane alignment. The full-width-half-maximum (FWHM) of the
curve is 5.4 55 5 ........................................................................................................ 4 1

2-10. The x-ray diffraction 4-scan of ( 11) Ni which indicates the substrate's in-plane
alignment. The average full-width-half-maximum (FWHM) of the 4 peaks is
7 .7 7 6 20 ..................................................................................................................... 4 2

2-11. The log-scale (111) Ni pole figure of the Ni substrate formed by RABiTS process.
The cube fraction of the substrate is 98.2694% .................................................. 43









2-12. The X-ray diffraction 0-20 scan graphs of textured Cu tapes formed by RABiTS
process (a) before annealing and (b) after annealing at 800'C in vacuum for 2
h o u rs ......................................................................................................................... 4 4

2-13. The log-scale pole figures of(1 11) Cu substrates (a) before annealing and (b) after
annealing at 800'C in vacuum for two hours ...................................................... 45

2-14. Schematic vertical structure drawing of conventional 2G wire based on YBCO
H T S . ......................................................................................................................... 4 6

2-15. SEM cross-section of AMSC's 2G wire ............................................................. 46

2-16. Resistivity comparison of Cu tape and Ni-W (3 at.%) tape as a function of the
tem peratu re ............................................................................................................... 4 7

2-17. Simulated graph of cap layer thickness as a function of current per tape width ...... 48 2-18. Simulated graph of engineering current density as a function of current per tape
w id th ......................................................................................................................... 4 9

2-19. Thermodynamic stability curve of Cu in terms of temperature and oxygen partial
p re ssu re ..................................................................................................................... 5 0

2-20. Schematic drawing of magnetron sputtering system ........................................... 51

2-2 1. Schematic drawing of electron beam evaporation system ................................... 52

2-22. Schematic drawing of pulsed laser deposition system ........................................ 53

2-23. Schematic drawing of typical perovskite crystal structure ................................. 54

2-24. Schematic diagram of Zaanen, Sawatzky and Allen (ZSA) framework ............. 54

3-1. The vertical structure which is used for this experiment ...................................... 60

3-2. The X-ray diffraction 0-20 scan along the surface normal for a multilayer structure
annealed at 400'C ............................................................................................... 6 1

3-3. The X-ray diffraction rocking curve for the (004) Cu2Mg peak, indicating a fullwidth half-maximum (FWHM) of AO = 2.00 ..................................................... 62

3-4. The X-ray diffraction 4-scan through the Cu2Mg (222) ........................................ 63

3-5. The X-ray diffraction 0-20 scan of the (Cu,Mg) multilayer after annealing at 400,
500, 600, 7000C .................................................................................................. . . 64

3-6. The SEM picture of(a) CeO2 fin on Ni / Cu / MgO (b) CeO2 / Ni / (Cu,Mg) / Cu /
M gO structure ................................................................................................... . . 65








4-1. The X-ray diffraction 0-20 scan of (La,Sr) li03 film grown on SrTi03 single crystal
substrate by pulsed laser deposition (PLD) method at 750'C, in vacuum ....... 74

4-2. The in situ reflection high-energy electron diffraction (RHEED) pattern of
(LaSr)TiO3 film deposited on SrTiO3 single crystal substrate by PLD method at
750�C, in vacuum.RHEED is used to monitor the epitaxial film growth ........... 75

4-3. The X-ray 0-20 scans of (La,Sr)TiO3 films deposited on SrTiO3 single crystal
substrates in different ambient conditions .......................................................... 76

4-4. The X-ray diffraction 0-20 scan of LaTiO3 film grown on SrTiOj single crystal
substrate by pulsed laser deposition (PLD) method at 750'C, in vacuum ....... 77

4-5. The two different X-ray 0-20 scans of LaTiO3 film grown on SrTi03 single crystal
substrate in the oxygen pressure of 3 10-5 Torr. aligned to (a) the substrate (00 1)
plane and (b) the LaTi'O7 (-210) plane, respectively ....................................... 78
4-6. The two different X-ray 0-20 scans of LaTiO3 film grown on SrTiO; single crystal

substrate in the oxygen pressure of 4.Ox 104 Torr. aligned to (a) the substrate (001)
plane and (b) the La2Ti207 (-2 10) plane, respectively ........................................ 79

4-7. The X-ray k-scan of (-420) peak for La2Ti207 layer grown on SrTiO; single crystal
substrate in the oxygen pressure of 3.N 10-5 Torr ............................................... 80

4-8. The resistivities of (LaSr)IiO3 and LaTiO3 films on single crystal SrTiO3. measured
at (a) 300 K and (b) 77 K, as a function of the oxygen pressure ........................ 81

4-9. The resistivity curves of(La,Sr)TiO3 films on SrTiO3 substrates grown in (a)
vacuum, (b) 3.N I0 Torr of oxygen, and (c) 4.Ox 104 Torr of oxygen. as a function of tem perature ................................................................................................... . . 82

4-10. The resistivity curves of LaTiOG films on SrTiO3 substrates grown in (a) vacuum.
(b) 3.0 10-5 Torr of oxygen. and (c) 4.Ox10-4 Torr of oxygen, as a function of
tem peratu re ............................................................................................................... 8 3

4-11. The X-ray 0-20 scan of YBa2Cu307 deposited on (LaSr)Ti03 buffer layer on
SrTiO 3 single crystal substrate .......................................................................... 84

4-12. The X-ray o)-scan of(a) (002) peak for (L.a.Sr)TiO3 buffer layer on SrTiO3 single
crystal substrate and (b) (006) peak for the YBaICu307 film deposited on
(La,Sr)T iO 3 buffer layer ...................................................................................... 85

4-13. The X-ray 4-scan of(a) (112) peak for (La.Sr)TiO3 buffer layer on SrTiO3 single
crystal substrate and (b) (012) peak for the YBa2Cu307 film deposited on
(La.Sr)TiO 3 buffer layer ...................................................................................... 86








4-14. The resistivity versus temperature measurement for YBa2Cu307 film deposited on
(La,Sr)TiO3 buffer layer on single crystal SrTiO3 substrate ............................... 87

4-15. The critical current density (Jc) as a function of magnetic field of YBa2Cu307 film
deposited on (La.Sr)TiO3 buffer layer on single crystal SrTiO3 substrate .......... 88

4-16. The resistivity versus temp. graph of (La,Sr)TiO3 film grown on single crystal
SrTiO3 substrate after annealing with the YBa2Cu307 deposition condition ..... 89

5-1. The X-ray diffraction 0-20 scan of (La,Sr)TiO3 films deposited directly on Ni-W
tape by PLD at the temperature of (a) 700'C, (b) 750�C. and (c) 800'C ........... 97

5-2. The energy dispersive X-ray spectroscopy (EDS) results of TiN films grown on Cu
layers on single crystal SrTiO3 substrates with the thicknesses of (a) 500A. (b) 1 000A, (c) 2000A, and (d) 5000A. These curves were taken after annealing the
samples at 740'C, in vacuum for 60 minutes ...................................................... 98

5-3. The atomic percent of Cu observed on the surface of TiN layer with the thickness
range of 500-5000A. The square symbols designate the samples without annealing.
The circles and triangles show the samples with annealing at 740'C, in vacuum for 30 minutes and 60 minutes, respectively ............................................................. 99

5-4. The X-ray diffraction 0-20 scan of (a) TiN layer deposited on textured Ni-W alloy
tape, and (b) (La,Sr)TiO3 film deposited on TiN seed layer .................................. 100

5-5. The in situ reflection high-energy electron diffraction (RHEED) pattern of (a) TiN
film deposited on textured Ni-W alloy tape, and (b) (La,Sr)TiO3 film deposited on
T iN seed layer ........................................................................................................ 10 1

5-6. The X-ray diffraction o)-scan of (a) Ni-W (002). (b) TiN (002). and (c) (La,Sr)TiO3
(0 04 ) p lanes ............................................................................................................ 102

5-7. The X-ray diffraction 4-scan of(a) Ni-W (I11), (b) TiN (11), and (c) (LaSr)TiO3
(1 12 ) p lan es ............................................................................................................ 10 3

5-8. The X-ray diffraction 0-20 scan of high temperature superconducting YBa2Cu307
film grown on (La,Sr)TiO3 buffer layer / TiN seed layer / Ni-W tape ................... 104

5-9. The X-ray diffraction (,)-scan of (005) YBa2Cu307 which was grown on (La.Sr)TiO3
buffer layer / TiN seed layer / N i-W tape ............................................................... 105

5-10. The X-ray diffraction 4-scan of (103) YBa2Cu307 which was grown on (LaSr)TiO3
buffer layer / TiN seed layer / N i-W tape ............................................................... 106

5-11. The resistivity versus temperature graph for YBazCu307 film grown on (LaSr)TiO3
buffer layer / TiN seed layer / N i-W tape ............................................................... 107








5-12. The voltage versus current graph of YBa2Cu307 film deposited on (La,Sr)TiO3
buffer layer / TiN seed layer / N i-W tape ............................................................... 108

5-13. The SEM picture of YBa2Cu307 film surface deposited on (La,Sr)TiO3 buffer layer
/ T iN seed layer / N i-W tape .................................................................................. 109

5-14. The energy dispersive X-ray spectroscopy (EDS) results of YBa2Cu307 film grown
on (La,Sr)TiO 3 / TiN / N i-W tape ......................................................................... 110

6-1. The surface SEM picture of the (La,Sr)TiO3 film grown on Cu tape after annealing at
780'C in oxygen partial pressure of I.0x 10-1 Torr for 7 minutes .......................... 118

6-2. The energy dispersive X-ray spectroscopy (EDS) graph of (La,Sr)TiO3 film grown
on Cu tape after annealing at 780'C in oxygen I.Ox 10- Torr for 7 minutes ......... 119

6-3. The X-ray diffraction 0-20 scan of(a) Ir layer deposited on textured Cu by PLD, and
(b) (La.Sr)TiO3 film deposited on Ir layer ............................................................. 120

6-4. The in situ reflection high-energy electron diffraction (RIJEED) pattern of Ir film
deposited on textured Cu tape, and (b) (La,Sr)TiO3 film deposited on Ir layer..... 121 6-5. The X-ray diffraction (,)-scan of(a) Cu (002), (b) Ir (002), and (c) (La,Sr)TiO3 (004)
p lan e s ...................................................................................................................... 12 2

6-6. The X-ray diffraction f-scan of (LaSr)TiO3 (112) which was grown on Ir film on Cu
ta p e . ........................................................................................................................ 12 3

6-7. The X-ray pole figures of (a) Cu (111), (b) Jr (111), and (c) (LaSr)TiO3 (112).....124 6-8. The resistivity curves of Ir on SrTiO3 single crystal, (La,Sr)TiO3 film on SrTiO3
substrate and (1La,Sr)TiO3 film on Ir on SrTiO; substrate ...................................... 125

6-9. The resistivity curves of Ir on SrTiO3 single crystal with and without annealing at
780'C in oxygen 1.0X104 Torr tbr 7 m inutes ........................................................ 126

6-10. The X-ray diffraction 0-20 scan of YBa2Cu3O7 layer deposited on (La,Sr)TiO3 / Ir
m ulti buffer stack on textured Cu tape ................................................................... 127

6-11. The surface SEM picture of the YBau307 film grown on (La,Sr)TiO3/ Ir buffer
stack on Cu tape with the magnification of(a) x l000 and (b) x5000 .................... 128

6-12. The energy dispersive X-ray spectroscopy (EDS) graph of (a) normal YBa2Cu307
film surface, (b) defect region of YBa2Cu307 film ................................................ 129

6-13. The resistivity versus temperature measurement for YBa2Cu307 film deposited on
(La,Sr)TiO 3 / Ir m ulti buffer stack on Cu tape ....................................................... 130













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

NUCLEATION AND EPITAXY OF CONDUCTIVE BUFFERS ON (001) Cu FOR
COATED HIGH-TEMPERATURE SUPERCONDUCTING CONDUCTORS By

Kyunghoon Kim

August 2005

Chair: David P. Norton
Major Department: Materials Science and Engineering

In the 2nd generation wire technology of high temperature superconducting coated conductors, highly conductive and nonmagnetic Cu substrate can improve the wire properties along with the conductive buffer layers, offering fully conductive wire architecture. This scheme requires two components, namely oxidation resistance for the Cu tape and conductive buffer layers.

The growth of epitaxial Cu2Mg as an oxidation barrier was investigated. Epitaxy of (004) Cu2Mg intermetallic phase was achieved on (002) Cu film. An in-plane qD-scan through the Cu2Mg (222) and the x-ray diffraction rocking curve for the (004) Cu2Mg peak indicates that the intermetallic phase is well oriented on (002) Cu surface.

The perovskite (LaSr)TiO3 was investigated as a possible conducting oxide buffer layer for high temperature superconducting coated conductors. YBa2Cu3O7 was grown epitaxially on (LaSr)TiO3 buffer layer on SrTiO3 substrate with excellent in-plane and out-of-plane alignment. The superconducting transition temperature (T,) of YBa2Cu307 /








(La,Sr)TiO3 / SrTiO3 structure was 91 K and the critical current density (Jc) of this structure was 2.18 x 106 A/cm2 at 0 magnetic field. The resistivity results of a post annealed sample at YBa2Cu307 deposition condition indicates that the (LaSr)TiO3 layer can be a candidate for the conductive buffer layer in the coated conductor applications.

The epitaxial film growth of (LaSr)TiO3 was examined on Ni-W metal alloy tape. The transition metal nitride such as TiN was deposited epitaxially on Ni-W tape by PLD and played an excellent role as a seed layer for (LaSr)TiO3 film growth on Ni-W tape. The YBa2Cu307 film was deposited epitaxially on the (LaSr)TiO3 buffer layer with the TiN seed layer on Ni-W tape. The YBa2Cu307 film grown on (LaSr)TiO3 / TiN / Ni-W tape has T. of 89 K and J, of 0.42x 106 A/cm2.

The epitaxial film growth of (LaSr)TiO3 was examined on Cu tape as a possible conducting buffer layer for high temperature superconducting coated conductors. The noble metal, such as Ir, was deposited epitaxially on Cu tape by PLD for an oxygen diffusion barrier. The YBa2Cu307 film was deposited epitaxially on the (LaSr)TiO3 and Ir buffer stack on Cu tape. The YBa2Cu307 film grown on the (La,Sr)TiO3 / Ir / Cu tape has a superconducting transition temperature of 90 K and a critical current density value of 1.Ox 106 A/cm2. This shows that (LaSr)TiO3 is a possible candidate for the conductive buffer layer in the Cu based RABiTS applications.













CHAPTER 1
INTRODUCTION

The coated high-temperature superconducting (HTS) conductors consist of metal substrate, buffer layers and thin film superconducting oxide. The rolling-assisted biaxially textured substrate (RABiTS) process made it possible to offer long length HTS wire. The advantages of Cu tape with nonmagnetic, lower material cost (-20% of Ni), easy formation of a sharp cube texture and higher electrical and thermal conductivity (for Cu, k: 398 W/m-K and for Ni, k: 90 W/m-K) than Ni or Ni based alloy tapes can improve the YBa2Cu307 based 2G wire properties along with the conductive buffer layers, which offers fully conductive wire architecture.

Though the Cu based RABiTS process with the conductive buffer architecture has advantages, the Cu oxidation problem could be worse. In metallurgical applications, enhancement of Cu oxidation resistance has been investigated by several approaches. Among these, the use of Mg-doped Cu films as an oxygen diffusion barrier is being investigated with the intermetallic phase of Cu2Mg.

There are key issues for a conductive buffer layer of HTS coated conductors. First, it must be reasonably well lattice matched to both the metal substrate and the superconducting film, thus enabling epitaxy. Second, the interaction between the buffer layer and the metal substrate must be such as to minimize formation of any native interfacial oxide that would serve as an insulating barrier to shunted current flow. An alternative candidate material system that may satisfy the criteria for conductive buffers mentioned above is (LaSr)TiO3. LaTiO3+x is an interesting defect perovskite system, with








transport properties varying from insulating to metallic based on oxygen stoichiometry. La and Ti have a relatively high affinity for oxygen. This suggests that the driving force for native oxide formation at the interface between LaTiO3 and either Ni or Cu should be reasonably low. With an extreme sensitivity to oxygen content, LaTiO3+, is not particularly attractive as a conductive buffer layer. One possible approach to maintaining metallic conductivity in an oxidized state is through cation doping. This could maintain carrier density as well as reduce oxygen diffusivity. The most likely dopant candidate is Sr. Doping with a divalent element increases the Ti+3 / Ti4 ratio and can make the compound less sensitive to ambient oxygen pressure during YBCO deposition.

One objective of this work is to investigate the epitaxial growth of intermetallic

Cu2Mg phase as an oxygen diffusion barrier for high temperature superconducting coated conductors. The other one is to grow epitaxial (LaSr)TiO3 film as a conductive buffer layer for Cu based RABiTS applications.

Chapter 2 reviews the fundamental background of the high temperature

superconductors and the rolling-assisted biaxially textured substrate (RABiTS) process including textured metal substrates and buffer layers. In this chapter, preventing method of Cu oxidation, thin film techniques for oxides growth and overview of various conductive oxides for HTS coated conductors will be mentioned. Chapter 3 gives the results of epitaxial Cu2Mg growth as an oxygen diffusion barrier for Cu RABiTS. Chapter 4 describes the fundamental characteristics of the (LaSr)TiO3 film as a conductive buffer layer for coated conductors. The oxygen sensitivity of (LaSr)TiO3 film will be investigated with the comparison of LaTiO3 film. The electrical transport property such as resistivity as a function of temperature will be covered with various oxygen








pressure conditions. Chapter 5 is for the application of (LaSr)TiO3 film as a conductive buffer layer in Ni-W based RABiTS process. In this chapter, epitaxial TiN layer deposited by PLD method on Ni-W tape will be mentioned as a seed layer for (La,Sr)TiO3 film. Chapter 6 is for the application of (LaSr)TiO3 film as a conductive buffer layer in Cu based RABiTS architecture. Epitaxial Ir layer grown by PLD method on Cu tape will be described as an oxygen diffusion barrier. From the result of chapter 6, the feasibility of (LaSr)TiO3 film as a conductive buffer layer for Cu based coated conductor will be confirmed. Finally, chapter 7 is the summary for my dissertation.













CHAPTER 2
LITERATURE REVIEW

2.1 High Temperature Superconductors

Superconductivity is an electronic phase transition in which a metal displays a dc resistivity of identically zero and behaves as a perfect diamagnetic material in excluding magnetic field. Most metals do not exhibit a superconducting phase. The low temperature superconductors (LTS) show zero electrical resistance usually well below 20 K. Single elements such as Hg (superconducting transition temperature T, : 4.1 K), Pb (Tc : 7.2 K), Nb (T, : 9.2 K) and various compounds show superconducting behavior at low temperatures. Nb3Sn (Tc : 18.1 K) known as Al 5 superconductors [1], Lil+xTi2-xO4 (Te: 13.7 K) [2] and BaPbl,BixO3 (Tc : 11.7 K) [3] are the examples of compound LTS.

The BCS theory (proposed by John Bardeen, Leon Cooper and Robert Shrieffer in 1957) gives an explanation of superconductivity behavior [4]. According to this model, pairs of electrons (Cooper pair) interact with crystal lattice vibrations in such as way so that an attractive potential results from this interaction. The electron pairs have a slightly lower energy than unpaired electrons and produce an energy gap which decreases from about 3.5kB'Tc at T=0 K (where kB is Boltzmann constant = 8.616x10-5 eV/ K) to zero at Tc. This reduced energy gap as a function of temperature was experimentally supported by Townsend et al. [5]. Figure 2-1 is schematic drawing of the energy gap of superconductor material and the energy gap dependence on temperature with theoretical value vs experimental results for some elemental metals [6]. There are two important characteristic lengths associated with superconductivity. One is the London penetration








depth, which describes the magnetic field decay inside the superconductor [7]. The penetration depth, L, is defined by B, - Bz(O) exp(-x/L), where B. is the magnetic field inside the superconducting material, Bz(O) is the magnetic field at the interface and x is the distance from the interface. The exclusion of magnetic field by superconductors, known as the Meissner effect, can be explained by this London theory. The other characteristic length is the coherence length, 4, which is related to the transition layer thickness from superconducting state to a normal state. The coherence length is the decay distance of the superconducting wavefunction [8].

In addition to Tc, there are two more critical parameters that characterize

superconducting behavior. The critical magnetic field, Hc, is the magnetic field strength above which the superconductivity disappears. It depends on the temperature as described by Hc(T) = H(O) [1 - (T / To)2]. Superconductivity also vanishes if too much current is flowing through the material. The limiting current is called the critical current Ic. Superconductivity remains only when the temperature, magnetic field and current are below these three limiting factors (T,, H,, I). Figure 2-2 shows the phase space within which the superconductivity remains (T-H-I diagram). A sharp transition from the superconducting state to a normal state with applied magnetic field is observed in type I superconductors. Practical limitation exists in Type I superconductor because of the low H,, value. Type II superconductors has somewhat different transition behavior. The transition from the superconducting state to a normal state is gradual, and there exists an upper critical magnetic field H.2 which is higher than H.1. The type I and II superconductors can be distinguished by using the Ginzburg-Landau parameter, K, which is the ratio of penetration depth XL to coherence length (K XL / g). A Type I









superconductor shows the K value far below 1 (4 > XL). Pure, elemental superconductors are in this category. When XL is larger than 4 (K 1), such superconductors are in the type II category.

The magnetic field begins to penetrate through the type II superconductors when the field strength reaches a lower critical magnetic field (H.1). This phenomenon is described as a vortex state. Type II superconductors have cylindrically symmetric domains called vortices which are in the normal state and surrounded by the superconducting matrix. An external magnetic field can penetrate through these vortices which the material remains in the superconducting state.A super conducting current can be maintained in the superconducting matrix if the vortices are pinned to their positions (fluxoid pinning). As the temperature or the magnetic field is increased, the vortex density increases, with vortices getting close to each other. Finally the superconductivity disappears at the upper critical magnetic field (H.2). Figure 2-3 shows the vortex state or mixed state which exists in the type II superconductors. Superconducting alloys and compounds show type II superconductivity with relatively high transition temperatures, flowing large currents and often operating in large magnetic fields [9]. Type 11 superconductors are sometimes referred to as hard superconductors because of these properties mentioned.

A high temperature superconductor (HTS) typically describes a material in which the superconducting transition temperature is greater than 30 K. In 1986, Bednorz and Miller discovered Ba-La-Cu-O system which showed superconductivity in the 30 K range [10]. Various oxide compounds have been explored since this first HTS discovery. Y-Ba-Cu-O compounds showing superconductivity transition between 80 and 93 K were








reported by Wu et al. in 1987 [11]. Maeda et al. discovered the Bi-Sr-Ca-Cu-O system with the Tc of about 105 K in 1988 [12]. TheTl-Ba-Ca-Cu-O superconductor, which showed 120 K transition temperature, was discovered by Sheng et al. in 1988 [13]. The highest T, (above 120 K) was reported in the Hg-Ba-Ca-Cu-O system by Putilin et al. in 1993 [14].

The generic structure of all HTS compounds consists of layered CuO2 planes and charge reservoir blocks in a unit lattice cell [15]. The CuO2 planes are separated by divalent or trivalent atoms. The superconductivity, as well as charge transport, are mostly confined to the CuO2 planes. Figure 2-4 is the schematic drawing of La2CuO4. Carrier doping by substitution of alkaline earth atoms with trivalent rare earth atoms such as (La,Sr)2CuO4 makes it superconducting at 40 K. YBa2Cu307 has been extensively explored for superconducting devices and wires. YBa2Cu307-8 is a hole-doped superconductor possessing a Cu3 / Cu2+ valence state mixture. Its crystal structure is orthorhombic with a = 3.82A, b = 3.88A, and c = 11.68A [16]. Figure 2-5 is the schematic drawing of the YBa2Cu307-8 unit cell which can be thought of as an oxygen deficient perovskite structure. One unit cell of YBa2Cu3OT7 contains one Y atom, two Ba atoms, three Cu atoms and seven 0 atoms. The YBa2Cu307-8 unit cell consists of two CuO2 planes separated by a Y atom. CuO chains are between the Ba-O layers. Growth temperature and oxygen stoichiometry affects the oxygen site occupancy and transition temperature of the YBa2Cu307-8 compound. For films, the lattice parameters can depend on growth temperature. The Tc can be reduced below 40 K with 8 = 0.7 [17, 18]. In the case of 8 = 1, YBa2Cu306 compound is not superconductor and shows a tetragonal structure with a = 3.857A and c = 11.81 9A [19]. Figure 2-6 is the schematic drawing of








YBa2Cu306 compound. The structural difference between 8 = 0 (Figure 2.5) and 8 1 (Figure 2-6) is in the CuO chains which play a crucial role in superconductivity. The highest J can be obtained for current flow parallel to the a-b planes, reflecting the anisotropic crystal structure characteristics. The c-axis oriented epitaxial YBa2Cu307 films can be deposited by various techniques. Coevaporation is one of the earliest proposed methods for YBa2Cu307 film growth [20-22]. Y, Cu, and BaF2 from separate sources are coevaporated and an ex situ annealing in oxygen ambient is performed. The RF magnetron sputtering method [23] and pulsed laser deposition method [24] for YBa2Cu307 film growth were also reported. The detailed thin film growth technique will be explained in Chapter 2.3.

The maximum superconducting transition temperature of the Bi-Sr-Ca-Cu-O

system was found at 105 K which is greater than that of YBa2Cu3O7. The Bi-Sr-Cu-O system without Ca has a maximum T value of 22 K [25,26]. The crystal structure of BiSr-Ca-Cu-O system is an incommensurate superstructure based on an orthorhombic subeell with a = 5.414A, b = 5.44A, c = 30.78A [27]. Figure 2-7 is the schematic drawings of Bi-Sr-Cu-O and Bi-Sr-Ca-Cu-O compounds. If there is no calcium [Figure 27(a)], the compound shows low temperature superconductivity. The important difference between Figure. 2-7 (b) and (c) is the number of Cu-O sheets. The compounds of double Cu-O sheets have orthorhombic c-axis of about 30A, and the compounds that have triple Cu-O sheets show c-axis of about 37A [28, 29]. This structural difference generated from the number of Cu-O sheets leads to the transition temperature alteration. The compound with the double Cu-O sheets shows a lower T, phase of around 80 K, but the compound with triple Cu-O sheets shows a higher Tc phase as shown in Maeda's R-T curve [12].








The three basic high temperature superconductors such as La-A-Cu-O (A : Ba, Sr, Ca), R-Ba-Cu-O (R: rare earth element), and Bi-Sr-Ca-Cu-O compound have Cu-O sheets where Cu is in essentially square-planar coordination with Cu2 . Subramanian suggested that the superconducting mechanism of the Bi-Sr-Ca-Cu-O compound is related to the orthorhombic distortion of pseudo-tetragonal sheets. Lowering the symmetry of the copper-oxygen sheets and the Bi-O sheets may play a role in this distortion [30]. The bulk Bi-Sr-Ca-Cu-O compounds can be synthesized by the procedure of calcining, grinding, pressing and sintering in air or oxygen ambient. Bismuth oxide (Bi203), strontium carbonate (SrCO3), calcium carbonate (CaCO3) and copper oxide (CuO) mixture can be used as the synthesizing agents. In order to produce Bi-Sr-Ca-Cu-O thin films for various device applications, laser sputtering [29], laser ablation [31 ], successive laser ablation with N20 gas [28] and laser molecular beam epitaxy [32] methods have been used. Lead (Pb) can be added to this compound to reduce the processing temperature and increase the transition temperature. In the study of the new cuprate superconductors, Kawai et al. observed that the basic structure units in Bi2Sr2Can. ICunO2n 4 and Tl2Ba2Can_CunO2n-4 consist of layers of Ca(Sr)CuO2 and n--1 compound [32]. For example, the mixture of CaCuO2 with Bi2Sr2CuO6 (nil) gives the compound of Bi2Sr2CaCu2Og (n=2, so-called Bi-2212), and the mixture of 2(CaCuO2) with Bi2Sr2CuO6 (n1l) gives Bi2Sr2Ca2Cu3Oj0 (n=3, so-called Bi-2223). For this reason, CaCuO2 is considered the parent structure of high temperature superconductors.

2.2 Superconducting Power Applications using HTS Wire

High-current power transmission cables are among the most exciting opportunities in the high temperature superconductor applications. The operating temperature of HTS materials such as Y-Ba-Cu-O and Bi-Pb-Sr-Ca-Cu-O compounds at liquid nitrogen of 77








K certainly has the attractive advantages. The cost savings of the liquid nitrogen is at least 50 times compared to liquid helium which is used to cool the LTS materials, such as Nb3Sn and NbTi, to 4.2 K. In addition, the cryogenic cooling and vacuum insulation systems will be simple by using liquid nitrogen. Other application examples of the superconductors include the field magnet for the motor, the rotor coil for a generator, transformer for power grids, a fault-current limiter which protects a power transmission and distribution system from surges, a current lead for reducing the heat loss in cryogenic machine, magnetic bearings which would be used with large flywheels, magnetically levitated trains for transportation, magnetic resonance imaging (MRI) for medical diagnostic instrument, and superconducting quantum interference devices (SQUID) for magnetic field sensors [33].

First-generation (I G) HTS wire using Bi-Pb-Sr-Ca-Cu-O compounds has the

disadvantage of high cost because it is manufactured with superconductive filaments in a silver matrix using the powder-in-tube (PIT) procedure. The detail process of PIT will be mentioned in Chapter 2.3. In order to produce cost-effective long length wire, the process using superconducting Y-Ba-Cu-O material grown on metal tape has been proposed which is known as the second-generation (2G) wire.

An important factor that can influence the current density, J,, in superconducting wire applications is the crystallinity of the HTS material. The high-angle grain boundaries generated in polycrystalline HTS reduce the critical current. The grain boundary can act as a weak superconducting interface. These are known as weak links. The grain boundary effect was demonstrated by Chaudhari et al. [34]. For this reason, the method of epitaxial YBa2Cu307 film deposition on biaxially textured metal tapes was








introduced [35]. The YBa2Cu307 film grown on textured metal templates can drastically reduce the misorientation of the individual grains allowing improvement of the links in the current path. The metal tapes produced by the thermomechanical texturing are known as rolling assisted biaxially textured substrate (RABiTS). The RABiTS process includes depositing buffer layers and HTS materials on the biaxially textured flexible metal substrates.

2.2.1 Biaxially Textured Metal Substrates

The basic procedure for RABiTS consists of cold rolling a metal bar in long lengths and subsequent annealing in reducing condition. The primary metals of interest are Ni, Cu and the alloys. The { 100}(100) cube texture of the metal tape has cube plane parallel to the plane of the sheet and a cube edge parallel to the rolling direction. By continuous rolling, a smooth surface of the metal substrate can be obtained with root mean square (rms) roughness of -50 nm [36]. Subsequent annealing of the as-rolled metal tape enables the sharp { 100}(100) cube texture. The annealing is performed at 800-1000�C in vacuum or ambient of Ar/H2 mixture gas for a few hours. Figure 2-8 shows the conceptual drawing of the RABiTS process including the buffer layer deposition [36].

The degree of texturing in metal substrates can be measured by x-ray diffraction. Figure 2-9 shows the x-ray o0-scan of the (002) Ni which indicates the Ni substrate formed by RABiTS process has out-of-plane alignment. This measurement is done by fixing the sample in the 20 position of (002) Ni and scaning the 0 angle. Figure 2-10 shows the x-ray +-scan of the (111) Ni which indicates the substrate has in-plane alignment. This measurement is done by fixing the sample in the 20 and Y position of (111) Ni, and scanning the + angle. Scaninng the + angle is necessary to observe the in-









plane-alignment. The 20 angle of (002) Ni is 51.8440 [37]. The w angle of( 111) Ni is 350 because the Ni has face-centered-cubic (fcc) crystal structure. The full-width-halfmaximum (FWHM) of o- and O-scan graph is calculated by fitting Gaussian distribution to the measured data points. For the sample shown, the FWHM values of o- and 4-scan are 5.4555' and 7.7762', respectively. These are well matched with the typical data of Ni based RABiTS [38]. Figure 2-11 is the log-scale (111) Ni pole figure of the Ni tape. The pole figure is obtained by fixing the 20 angle of(1 11) Ni and rotating the sample by 3600 at an individual y angle from 0' to 900. The cube fraction of textured substrate can be calculated by summing the intensities at the cube orientation locations and dividing by the total integrated intensity in the log-scale pole figure. The cube fraction for the sample in figure 2.11 is 98.2694%. The four dark circles indicate a well-developed, single component cube texture. The influence of the annealing on the cube texture can also be observed by the x-ray diffraction method. Figure 2-12 shows the x-ray diffraction 0-20 scan graphs of textured Cu substrates before and after annealing. The annealing condition is 800�C in vacuum for 2 hours. The 0-20 scan can not tell any differences between the before and after annealing samples because it only finds every 20 angles that satisfies the Bragg's condition (nX = 2d-sin0, here X is the wave length of incident x-ray, d is the inter-plane distance and 0 is the scattering angle). The 20 angle of (002) Cu is 50.43090 [37]. On the other hand, the pole figure graphs designate clear differences of cube texture between the two treatments. Figure 2-13 is the log-scale pole figure of (111) Cu formed by RABiTS process before and after annealing. Without annealing, there are many satellite intensities between each 4 peaks which indicates that the texture is not clearly aligned. After annealing at 800'C in vacuum for 2 hours, the satellite peaks disappear and








the cube fraction was 97.2784% which is higher than the sample without annealing of 84.6100%.

Pure Ni is ferromagnetic. This contributes to ac losses. Alloying Ni with W and Fe [39], W only [40] or Cr [41] were suggested to overcome this ferromagnetic problem. Addition of alloying element also increases the yield strength of the metal tape which can be acceptable for a large number of applications. Addition of 3 at. % W can enhance the yield strength of pure Ni from 34 MPa to 150 MPa and reduce the Curie temperature from 627 K to around 400 K [42].

2.2.2 Buffer Layers

The first generation of high temperature superconducting wire technology was led by powder in tube (PIT) process with Bi-Pb-Sr-Ca-Cu-O superconductor material. On the other hand, the second generation (2G) wire is being developed by the RABiTS process with Y-Ba-Cu-O material. The RABiTS process consists of a biaxially textured metal substrate such as Ni, Cu or their alloys, buffer layers and YBa2Cu307. The growth of HTS directly on the metal substrate has several obstacles. Due to the oxygen ambient ( 10-1 Torr) at elevated temperature ( - 800'C) during the HTS film formation, the metal substrate can be oxidized resulting in metal oxide on the metal surface. Another problem is the cation substitution in the HTS material. For these reasons, buffer layers between the metal substrate and HTS material have an important role. The buffer layers should prevent oxidation of metal substrate and metal diffusion through the HTS material during the high oxygen pressure and high temperature process. In the case of forming metal oxide on metal substrate, epitaxial layer growth can be impeded which affects the epitaxial layer formation of HTS film. As mentioned in the previous section, the realization of an in plane and out-of-plane aligned HTS film is crucial for achieving high








critical current density J. The metal oxide also affects the mechanical strength of final coated conductors. When selecting proper buffer layers in the RABiTS process, one should consider the oxidation of metal substrate during buffer layer deposition itself. This is because the process condition of buffer layers can be at high temperature and in an oxidizing ambient. The solution to this problem is to find stable buffer layer materials relative to NiO, CuO or Cu2O in the growth condition. Jackson et al. reported the thermodynamic stability curves of NiO, CuO and Cu2O compared to several oxides by calculating Gibbs free energies of the reactants and products [43]. According to this report, MgO, CeO2, Y203 are good possible candidates for buffer layers. Another important role of buffer layers is chemical separation of HTS from the cation contamination. YBa2Cu307 has proved to accommodate various cationic and anionic substitutions [44]. Among these, substitution of Cu with metallic elements dramatically affects the nature of the high-T, superconductivity. In the YBa2(Cu_..M,)307-8 system where M = Ti, Cr, Mn, Fe, Co, Ni, Ga and Zn, 3 - 10% of metal contamination can reduce the T. value below the boiling temperature of liquid nitrogen [45-49]. Table 2-1 shows the T, values reported for YBa2(Cui..M.)307_ systems. The third important aspect of buffer layers is the mechanical stability and proper adhesion to the metal substrate. For this reason, the lattice constant of the buffer layer should provide a reasonable match with both the metal substrate and HTS materials. The thermal expansion coefficient of the buffer layer also needs to be considered.

Buffer layer materials has also been studied for the microelectronic applications of superconducting devices such as superconductor-normal metal-superconductor (S-N-S) junctions or superconductor-insulator-superconductor (S-I-S) tunnel junctions. In these








applications, the starting substrates were single crystal materials such as Si, sapphire (A1203), GaAs, MgO or SrTiO3. It is difficult to grow high temperature superconducting thin films directly on these substrates because of interdiffusion and the lattice mismatch between them. Several buffer layers were explored for the superconducting microelectronic devices such as MgO [50-52], SrTiO3 [53-55], Y203 [56,57], Yttriastabilized zirconia (YSZ) [58-60], CeO2 [61], ZrO2 [62-64]. In the RABiTS application for 2G wire technique, YSZ / CeO2 multilayer buffer scheme on Ni tape was proposed by Norton et al. in 1996 [65]. In this architecture, the CeO2 layer is the epitaxial template, and YSZ has a role of alleviating oxide cracks. All of the multilayer buffer films were grown by pulsed laser deposition (PLD) method using a KrF excimer laser. After YBa2Cu307 formed on this buffer scheme, the superconducting transition temperature

(T,) was observed at 88 K and the critical current density (J,) for these initial RABiTS structures was 7x 105 A/cm2 at 77 K, 0 T. These results were comparable to those obtained for the epitaxial films on single crystal substrates. In order to retard metal oxidation effectively and overcome the slow deposition rate of PLD method, Pd deposition directly on Ni surface and electron beam evaporation deposition method for the multilayer buffer scheme were proposed. Goyal et al. obtained Jc value of 3 x 105 A/cm2 at 77 K, 0 T by using PLD deposited YBCO / YSZ / CeO2 buffer scheme on ebeam evaporated Pd on Ni tape [35]. Paranthaman et al. proposed e-beam evaporated YSZ / CeO2 buffer layers on Ni tape with crack free CeO2 thin film of 3-10 nm thickness [66]. Furthermore, He et al. studied YSZ / CeO2 buffer layers on Ni tape by using RF or DC magnetron sputtering method [67]. Mathis et al. could get superconducting properties of T. around 88 K, J, exceeding 3x 106 A/cm2 at 77 K, 0 T by adapting RF sputtered YSZ








and e-beam evaporated CeO2 multilayer scheme [68]. Another buffer scheme using transition metal nitride film was reported by Kim et al. in 2002 [69]. In this report, TiN was chosen because of low electrical resistivity (20-30 pQ-cm) and good mechanical strength (Young's modulus: 600 GPa, micro-hardness : 2000 Kg/mm2). Due to the oxidation of TiN layer during YBCO deposition, CeO2 layer also applied between YBCO and TiN films. The TiN layer was formed by DC reactive sputtering with Ar/N2 mixture gas and the CeO2, YBCO layers were grown by PLD method. A superconducting transition temperature for this architecture was 89 K and the critical current density was 6x105 A/cm2 at 77 K. Because TiN is also known as Cu diffusion barrier [70], Cantoni et al. studied multi buffer scheme using TiN for Cu based RABiTS process [71]. In this report, LaMnO3 / MgO / TiN buffer architecture was proposed. MgO layer was chosen for oxygen diffusion barrier, and LaMnO3 layer proved to be a planarizing material for smooth growing of YBCO. Table 2-2 shows the list of buffer layer materials tested with RABiTS applications referred from [42].

2.3 Review of Conducting Oxide Buffers for Cu based YBCO Coated Conductors

2.3.1 Introduction

Since the high temperature superconductors (HTS) were discovered in the late

1980s, worldwide efforts have been made to achieve high-efficiency electric wires. The first generation (I G) multifilamentary HTS wires are composites which can be produced with Bi2-.Pb.Sr2Ca2Cu3Olo (known as BSCCO-2223) and silver or silver alloy. BSCCO based HTS wire has shown a critical current density above 100,000 A/cm2 at 77 K [72]. Critical current density (Jc) is one of the limiting parameters of superconductivity, above which superconductivity disappears. Bi-Sr-Ca-Cu-O system has a high critical








temperature (T) value as high as 110 K with no rare earth elements [73]. The fabrication method of superconducting tapes with BSCCO and silver is known as powder-in-tube (PIT) process. The high-purity mixture of Bi203, PbO, SrCO3, CaO and CuO powder is filled into a silver tube and continuously drawn into a narrow cylinder of about lmm diameter. Such filaments are rolled together and deformed into a tape. Then a final annealing procedure is performed to react and obtain the tapes with good superconducting properties [74].

In order to overcome the high production cost - $300/kA.m - of IG wires,

YBa2Cu307_8 (YBCO) based second generation (2G) wires or tapes are widely being developed in both laboratories and industries [75]. 2G wires have the advantages of lower cost and better electrical performance under applied magnetic fields. The 2G wire with YBCO HTS on RABiTS architecture used by American Superconductor Corporation Inc. (AMSC) is made up of 3 buffer layers scheme on textured Ni-W (5 at.%) tape [76]. Y203 which can be formed by electron beam evaporation technique serves as a seed layer on 75pm thick Ni-W metal substrate. Yttrium-stabilized zirconia (YSZ) is deposited as a barrier layer. Due to the low level of lattice mismatch with YBCO (-0. 12%), CeO2 plays a role as a cap layer. Both YSZ and CeO2 layers are deposited by an if-sputtering method. Each buffer layer has a thickness of 75nm. The YBCO film, with the thickness of about 1 inm, is deposited on this buffer stack by metal organic decomposition (MOD) method using trifluoroacetate (TFA) based precursors. Silver is deposited on YBCO film with 3ptm thickness for capping the superconductor layer. Finally, a 75trm thick Cu film is bonded for mechanical and electrical stability. The Ag cap layer and Cu stabilizer can also serve as a current shunting path in the event of local defects existing in the YBCO








film. This shunting current cannot flow through the buffer layers because they are all insulators. Figure 2-14 is the schematic drawing of this YBCO based 2G wire architecture, and Figure 2-15 is the SEM cross-section of AMSC's 2G wire.

Nickel, which is the starting template of 2G wires, is ferromagnetic (FM) with a Curie temperature of 631 K and a saturation magnetization of 0.51 x 106 A/m at 0 K [77]. The magnetic metal substrate, such as Ni, can cause significant hysteretic losses during application of alternating current (ac). The ac loss can be decreased by adding W to Ni [78]. The Ni-W (5 at.%) alloy shows the Curie temperature of 339 K [79] and also increases the yield strength of the substrate to 165 MPa compared to pure Ni substrate of 34 MPa [78]. However, the Ni-W (5 at.%) alloy tape is not nonmagnetic at 77 K. In this aspect, Cu substrate is an attractive candidate for the starting template of 2G wires. Cu is a diamagnetic material that shows no ac loss phenomenon. Cu also surpasses Ni alloy in the electrical conductivity characteristics. At 300 K, the resistivity of Cu tape is 1.5 x 10-6 f2-cm, which is lower than Ni-W (3 at.%) alloy tape of 2.5x10-5 0-cm. At the temperature of liquid nitrogen 77 K, Cu tape and Ni-W (3 at.%) tape show the resistivity value of

2.Ox10"7 K2-cm and 1.7x10-5 -cm respectively. Figure 2-16 is the resistivity curves of Cu tape and Ni-W (3 at.%) alloy tape as a function of temperature. Cu tape has { 100}<100> cube texture and obtained from randomly oriented metal bars by cold-rolling, followed by an anneal in vacuum at 8001C for 1 h.

In this 2G wire architecture, the YBCO layer is located in the neutral axis between the 77-78ptm thick Ni alloy substrate with 3 buffer layers, and the 77-78pWn thick Ag cap layer with Cu stabilizer. The total thickness of the wire is slightly more than 1501im. For determining the critical current density (J), we divide the electrical current value by the








cross-sectional area of YBCO which carries the super current in its superconducting state. The overall engineering critical current density (JE) deals with the current value through the whole cross-sectional area of the wire including not only YBCO but also the substrate, buffer layers, cap layer and stabilizer. One way of increasing the JE value is to reduce the total wire thickness. This can be done by reducing the thicknesses of both the silver cap layer and copper stabilizer, which accounts for half of the total thickness [80]. In this case, the remaining problem is that we have to supply a shunting current path through the starting metal substrate. The current 2G wire structure consists of insulating buffer layers, which means that there is no shunting current path if there is no metallic capping layer on the YBCO film. In order to make the current flow through the metal substrate without capping layer, there must be a conductive buffer layers between the HTS layer and the metal substrate.

Compared to Ni or Ni based alloy tapes for 2G wire technology, several profitable aspects can be found in Cu substrates. Cantoni et al. have reported that the thermal stability of coated conductor has the relations with the capping layer thickness and the engineering current density [80]. Figure 2-17 and 2-18 are the simulated graphs of capping layer thickness and engineering current density as a function of the current-width values. The expressions for the capping layer thickness (d) and engineering current density (JE) are:

de = pc[(K2/w) - (ds]Ps + din/pm)],

JE = K/{ds + dm + pc[(K2/w) - (ds/ps + dm/Po)] }, where pc, ps, and Pm are the resistivities of capping layer (Pc : 2.x10-7 0-cm for Cu or Ag), superconducting film (ps : 5.0x 10-5 Q-cm), and metal substrate (Pm : 2.0x 10-7 Q.cm








for Cu and 4.5x 10-7 Q.cm for Ni). K is the current flow per tape width and w is the critical heat flux of the liquid nitrogen at 77 K (w : 10-20 W/cm2) [81]. The thicknesses of superconducting film and metal substrate are designated by d, (d, = K/J, Js : current density flown through the superconductor) and dm (din : 50pm). According to these graphs, no capping layer is needed up to K-500A/cm in the case of Cu tape and the maximum JE can be obtained up to -90 kA/cm2 for Cu tape which is higher than the value for Ni tape (-60 kA/cm2). The advantages of Cu tape (nonmagnetic, lower material cost (-20% of Ni), easy formation of a sharp cube texture [82] and higher electrical and thermal conductivity (for Cu, k : 398 W/m-K and for Ni, k : 90 W/m-K)) as compared to Ni or Ni based alloy tapes can improve the 2G wire properties along with the conductive buffer layers which offers fully conductive wire architecture. In this review, the possible problems of Cu based RABiTS process and overview of conductive buffer oxides will be addressed.

2.3.2 Preventing Methods of Cu Oxidation

In recent research, Aytug et al. demonstrated SrRuO3 (SRO) / LaNiO3 (LNO) as a conductive buffer structure for Ni based coated conductors [83]. Both SRO and LNO are perovskite type conductive oxides that show metallic behavior. The room temperature resistivity of SRO film is -1.8x 10-4 .cm and -3.0x 10-5 Q-cm at 4 K [84]. LNO has a resistivity of 1.8x 10-3 Q'cm at 290 K and 5x 10-4 0.cm at 4.2 K [85]. Pulsed laser deposition (PLD) method of YBCO on these conductive buffers is done at high temperatures (-780'C) and high oxygen partial pressures (-2x 10- Torr). These growth condition offer an oxidation environment to the metal substrate. From the work of Aytug et al., Ni substrate oxidation could be observed from XRD and cross-sectional SEM








analysis. Though the Cu based RABiTS process with the conductive buffer architecture has advantages, the oxidation problem could be worse. Figure 2-19 shows the thermodynamic stability curve which indicates that even -1 x 10-6 Torr of oxygen in the system at YBCO growing temperature can cause copper oxidation. Aytug et al. reported that La0.7Sr0.3MnO3 (LSMO) could be used as a conductive buffer layer on Cu based RABiTS applications [86]. In this work, they chose a Ni layer on metal substrate to prevent Cu oxidation. However, there are still remaining problems such as ferromagnetism of the Ni layer and NiO formation at the interface between substrate and buffer layer.

In metallurgical applications, enhancement of Cu oxidation resistance has been investigated by several approaches, including alloying or implanting elements such as Mg, Cr and Al that are known to be oxygen getters [87-92]. Among these, the Mg-Cu and Al-Cu systems are perhaps the most attractive due to the limited solubility of Cr in Cu [93]. The use of Mg-doped Cu films as an oxygen diffusion barrier is being investigated for Cu metallization in Si integrated circuit technologies [94]. Unfortunately, the resistivity of Cu increases rapidly with dopant concentration, deterring the use of an alloy as the RABiTS's substrate material if shunting functionality is desired. However, Mg-Cu or Al-Cu alloy thin film on the Cu substrate may prove to be suitable since the effective electrical transport path length would be limited to a thin surface layer.

Another method of preventing Cu oxidation is applying a noble metal as an oxygen barrier between metal substrate and conductive oxide buffer layers. One of the candidate material is iridium (Ir) which has face centered cubic (fcc) crystal structure and lattice parameter of 3.840A. In the crystallographic aspect, iridium can be well matched with








Cu substrate. Because iridium is a well known platinum group metal with excellent oxidation resistance, oxidation behavior is only considered above 1400'C [95-97]. Although the iridium oxides (IrO2) are formed after the YBCO deposition, it has good metallic property with resistivity below 3 x 10-4 0-cm at room temperature [98]. Paranthaman et al. first tried Ir as an oxygen barrier on Ni-W alloy based RABiTS application [99]. In their research, La0.7Sr0.3MnO3 conductive buffer layer was used to form fully conductive buffer architecture. On the other hand, careful consideration is also needed, because oxygen diffusivity through Ir is 5 x 1012 cm2/s at 800'C [78]. There is also a possibility of Cu diffusion through the Ir layer. Cu and Ir are known to be soluble in extremely small amounts [100]. However, 50gm thick Cu substrate can continuously supply copper element to the thin Ir layer during buffer oxides and YBCO processes. These mean that Cu oxidation can be observed either on top of the YBCO film or on the Cu substrate, and proper thickness of Ir layer should be deposited. Another candidate material for preventing Cu oxidation is Pd (palladium). Pd could be a good layer, but there is also an issue with miscibility in the Cu-Pd system.

2.3.3 Thin Film Techniques for Oxides Growth

In order to fabricate HTS coated conductors, not only the buffer oxides but also

superconducting oxides can be grown by various deposition techniques. These techniques include sputtering, electron beam evaporation, pulsed laser deposition (PLD), and nonvacuum deposition method known as chemical solution deposition (CSD).

Sputter deposition is a physical vapor deposition (PVD) technique that has been

used widely for thin film growth, especially in the semiconductor industries. Metal films, such as Al, Ti, and W, are sputter deposited for interconnection or metallization








applications. Usually Ar gas is introduced into the vacuum chamber (from few milli Torr to tens of milli Torr of pressure). The positively charged Ar ions generated by the potential difference between the cathode and anode electrodes bombard the negatively charged target materials. The ejected target elements from these collisions are deposited on the surface of the heated substrate materials. Radio frequency (RF) sputtering is the common way of depositing ceramic materials. Due to the phenomenon of impedance drop of dielectric materials in the high-frequency plasma, the current can flow through the dielectrics. A negatively self-biased target that results from a mobility difference between oscillating electrons and ions enables the insulating ceramic materials to be sputtered [101 ]. According to the Lorentz force, the path of each charged particle flowing one direction is bent by applying magnetic field. Electrons affected by magnetic field in the plasma move with spiral motion which is called as helix movement. This increases the ion density due to enhanced possibility of collision with gas elements. In other words, deposition pressure can be lowered maintaining the efficiency of sputtering yield. The deposition rate also can be improved with the same applied voltage. These are the advantages of magnetron sputtering. Figure 2-20 shows the schematic drawings of a planar type magnetron sputtering system. In the HTS coated conductor applications various buffer oxides such as YSZ [102], CeO2 [103], LaMnO3 [104], SrRuO3, LaNiO3 [83], Lao.7Sr03MnO3 [99] are epitaxially grown by the DC or RF sputtering methods. High deposition rate of sputtering system makes it useful for manufacturing long length wires.

Electron beam (e-beam) evaporation is one approach to evaporation deposition. In thermal evaporation, materials that needs to be deposited are placed in a crucible and








heated by a resistive heater. The vapor atoms are transferred to the substrates in the vacuum chamber at pressures typically less than 1.0xl0"3 Torr [105]. Due to the contamination problem of the crucible material at high temperatures, there is a limitation of using resistively heated evaporation sources. In the case of electron beam evaporation, contamination from the crucible and heating element can be drastically reduced by using shielded heating filaments [ 101]. The electrons thermionically emitted from this shielded heating filaments are deflected by a transverse magnetic field and reach the surface of the charged material. The evaporation starts by this electron beam heating of the source material. Figure 2-21 is a conceptual drawing of an electron beam evaporation system. Evaporation of MgO [106, 107], Y203 [108], CeO2 [109] are performed by the electron beam evaporation method for HTS buffer layers.

Pulsed laser deposition (PLD) is one of the commonly used methods for growing oxides. A KrF excimer is the most popular laser source [110]. A KrF excimer laser produces ultra violet (UV) light of 248nm wave length. The laser energy is absorbed by the target material and ablates the target atoms. The ejected atoms generate a plume and travel to the heated substrate material. The distance from the target to the substrate must be considered with respect to the background pressure and ambient gas pressure. Multilayered buffer oxides with various thickness range can be grown by shifting each target material without vacuum break and by changing the repetition rate. Figure 2-22 shows the schematic drawing of a simple PLD system. Due to its simplicity and ablating performance, most of the buffer oxides for HTS application can be deposited by PLD method. CeO2 [111], YSZ [112], MgO [113], LaMnO3 [113], TiN [113] buffer layers deposited by PLD for HTS application have been reported. The limited laser plume size








and the pulsed deposition mechanism can be a drawback of large scale fabrication of HTS coated conductors. However, high-rate pulsed laser deposition (HR-PLD) method has been proposed by Usoskin et al. [ 114].

Chemical solution deposition (CSD) under non-vacuum condition has great

advantages because of low cost and high deposition rate. The general procedures of CSD are synthesis of precursor solution, deposition by spin coating, formation of amorphous film by pyrolysis of organics and crystallization of the coating by high temperature heat treatment [115]. Sol-gel process using 2-methoxyethanol (CH3OCH2CH2OH) as a reagent of metal-oxygen-metal bond formation, and metal organic decomposition (MOD) process using carboxylate compounds as a metal cation source are the common CSD approaches. In the semiconductor industries, sol-gel process is widely used for deposition of TEOS (tetraethyl orthosilicate) oxide as an inter-metallic dielectrics (IMD). BaZrO3 [116], La2Zr207 [117], and CeO2 [118] buffer layers have been investigated by CSD method.

Finally, ion beam assisted deposition (IBAD) can be used in the case of

polycrystalline metal substrates in order to form the textured template for the buffer oxides. IBAD YSZ and IBAD MgO are the most common templates for producing 2G tapes [119].

2.3.4 Overview of Conductive Oxides for HTS Coated Conductors

An important aspect when considering epitaxial buffer oxides for HTS coated

conductors is lattice match with the c-axis aligned YBCO layer. Many of the conducting oxides are in the category of perovskite crystal structure possessing a pseudo-cubic lattice parameter which matches well with the a, b axes of YBCO (a=0.382nm, b=0.389nm).

The Perovskite structure consists of two metallic cations and three oxygen atoms (ABO3). The large A atoms occupy each corner of the lattice and oxygen atoms sit on








each face center with small B atoms in the lattice center, completing mostly orthorhombic crystal structure. From another frame of view, the small B atoms are surrounded by 6 oxygen atoms forming B03 (charge sharing notification) octahedron. This octahedron connects to other octahedra placing large A atom in the middle of their connection. This forms a pseudo-cubic structure. The A and B atoms are selected to make the charge balance. Figure 2-23 is the drawing of a typical perovskite lattice structure. The A site atoms can be cerium (Ce), calcium (Ca), sodium (Na), strontium (Sr), lead (Pb) and rare earth metals. The B site atoms can be titanium (Ti), niobium (Nb), iron (Fe), nickel (Ni), ruthenium (Ru), manganese (Mn), cobalt (Co), chromium (Cr) or copper (Cu) which are normally in the transition metal category.

The metallic, insulating and metal-insulator transition properties of a large number of perovskite oxide compounds have been explained by the framework model which includes correlation effects [120, 121]. The band widths Wof occupied oxygen 2p states and metal's d orbitals, the energy difference between oxygen 2p and the lowest unoccupied metal orbital which is designated as A, and the energy difference between the lowest unoccupied metal orbital and the highest occupied metal orbital which is designated as U' are the three terms that can be used in this model. Figure 2-24 is the schematic diagram of the Zaanen, Sawatzky and Allen (ZSA) framework. If the band width W is larger than A or U' as shown in the left and right end side of figure 2-24, the oxides are conductive. The left end side of figure 2-24 designates the overlap of the occupied 2p oxygen valence band with unoccupied metal conduction band, and the right end side designates the overlap of the occupied and unoccupied metal orbitals. In each case partially filled metal orbitals contribute to charge transfer. If U' is greater than A, the








oxide is called a charge-transfer insulator. If A is greater than U', the oxide is a MottHubbard insulator.

Lanthanum nickelate (LaNiO3) is a conductive perovskite oxide with pseudo-cubic lattice parameter of 0.383 nm. LaNiO3 is metallic due to the charge transfer gap A closing, which means that the occupied oxygen 2p orbitals and unoccupied 3d Ni orbitals are overlapped. A metal to insulator transition can be occurred in the oxygen deficient phase (LaNiO3.x) [122], and controlling of oxygen stoichiometry is important. Sputter deposited LaNiO3 buffered structure (YBCO / LaNiO3 / Ni tape) and multilayer buffer scheme (YBCO / SrRuO3 / LaNiO3 / Ni tape) are reported as conductive buffer oxides for Ni-based RABiTS [123, 83]. The SrRuO3 layer is used to prevent Ni diffusion through YBCO which can affect degradation of the Tc value.

The perovskite SrRuO3 has lattice constants of a = 0.555 inn, b = 0.556 rm, c =

0.786 rim, and a, b lattice constants are close to /2 ayBco or /2 bYBco with pseudo-cubic lattice parameter of 0.393 rin. The conductive property of SrRuO3 comes from the overlap between oxygen anion p orbitals and cation t2g [124]. Metallic SrRuO3 is also ferromagnetic with a Curie temperature of-160'K [125]. There have been several reports of SrRuO3 buffer layers as diffusion barriers on Pt [126] and Ni [127-129] for HTS coated conductor applications.

(Lail,,A,,)MnO3 is the interesting compound system which has both metallic and ferromagnetic properties. A represents a divalent alkali elements such as Sr, Ca or Ba. In the extreme composition of LaMnO3, La and Mn must have 3+ charges in accordance with the 6- charge of three oxygen atoms. LaMnO3 is an insulating antiferromagnet where electrons are localized on the atomic orbitals [130]. When the La atoms are








replaced by divalent elements in the concentration range of 0.2
LaTiO3 is an antiferromagnetic insulator (TN ; 150'K) when it has the fairly

stoichiometric composition with trivalent Ti3 ion. There exists a Mott-Hubbard energy gap Eg ;t U- W (see figure 2-24) which makes it an insulator. By changing the oxygen stoichiometry (LaTiO3+), a mixed valent state of Ti3+/ Ti4 can generate metallic transport properties [134, 135]. In the case of fully oxidized phase (LaTiO3.5 or La2Ti207), it becomes a ferroelectric band insulator. LaTiO3 has an orthorhombic perovskite structure with a = 5.604A, b = 5.595 A and c = 7.916 A [136]. In contrast with LaTiO3, La2Ti207 shows monoclinic layered structure. According to the phase diagram study, LaTiO3+x changes its electric and magnetic properties with oxygen content [137, 138]. LaTiO3+x is an attractive material for coated conductor application because it has good metallic property at all temperature ranges in the oxygen stoichiometric of 0.1 < x < 0.25 [137], and its pseudo-cubic lattice parameter is well matched with YBCO. However, the composition also can be changed during YBCO








growth in the high oxygen ambient (-I x 10"1 Torr) at high temperature (-780'C). The previous research on the relation of resistivity of LaTiO3+x with oxygen pressure during deposition by PLD indicated that oxygen played a crucial role in conducting property [139]. Cation doping in the compound can overcome this oxygen sensitivity of LaTiO3+x. Doping with divalent elements increases the Ti+3 / Ti4 ratio and can make the compound less sensitive to ambient oxygen pressure during YBCO deposition. Electrical conductivity is a function of doping constant x in the Lal.,SrTiO3 compound, and the resistivity continuously decreases with higher cation doping [140]. Previous studies on oxygen dependency showed that the room temperature resistivity of Lao.5Sro.5TiO3 remained below 1.0x0-3 Q-cm in 10-4 - 10-2 Torr of oxygen pressure range [141]. Recently, conductive buffer architecture with PLD deposited (LaSr)TiO3 film on Cu based RABiTS application (YBCO / LSTO / Ir / Cu tape) was reported [142].

La._xSrxCoO3 is the compound that has mixed valency due to the cation doping. The parent compound LaCoO3 is an insulator which has a charge transfer gap A (see figure 2.24) formed between an occupied oxygen 2p band and an empty cobalt 3d eg band [ 132]. By cation doping to LaCoO3, the ratio of Co3 /Co4+ increases introducing metallic behavior. The transport and magnetic properties depend on the doping concentration. The room temperature resistivity of Lao.5SrO.5CoO3 is known to be -9x 10-5 Q-cm. The pseudocubic lattice parameter of La0.5Sr0.5CoO3 is 3.835A. The oxygen stoichiometry is an important parameter for resistivity control because oxygen deficiency can change the Co3+/Co4+ ratio and the structural disorder in the Co-O-Co conduction channel [143].

LaCuO3 is a compound which has Cu3 ions. In a recent study, hybridization of unoccupied oxygen p states with Cu 3d and La 5d states was suggested as an electronic








structure of LaCuO3 [144]. The stoichiometric form of La2CuO4 is an insulator. However, the conduction behavior of LaCuO3_8 alters from an insulator to metallic conductor with 8 ranging from 0 to 0.5. Although the nonstoichiometric LaCuO3_8 has high conductivity with available oxygen vacancies, it is not a desirable material due to difficulties of synthesis under high oxygen partial pressure. Yu et al. reported that Sr doping in the LaCuO3_8 compound stabilized the perovskite structure. Enhancement of conductivity was observed in Lai-.xSrxCuO2- [145].

The overall electrical and structural properties of perovskite oxide compounds mentioned in this section is briefly shown in table 2-3. The major factor of selecting conductive buffer oxides for HTS coated conductor application should include not only the structural, chemical compatibility with metal substrate and YBCO film but also transport properties that need to be sustained even after the severe oxidizing process during YBCO formation.





31

Table 2-1. The critical temperature values in absolute temperature unit reported for
YBa2 CulM. 307-8 systems.

x Ti Cr Mn Fe Co Ni Zn Ga Ref. 0.1 75 84.5 78.9 38 21.2 66.3 < 3 45 0.033 61 55 80 70 51 46 0.1 86 92 50 50 73 40 47 0.033 76 73 55 55 48 0.033 88 88 80 58 49






32


Table 2-2. The list of buffer l er materials tested with RABiTS applications.
Cubic % Lattice Oye
Buffe% Lattice % Lattice Oxygen Electrical
Buffer Lattice Mismatch Diffusivity
Mismatch 2eitvt
Material Parameter versus (cm2/s) at
(A) YBCO versus Ni (C (Is.cm) (A BO800�C (jcm MgO 4.210 9.67 17.74 8x 10-22 SrTiO3 3.905 2.16 10.26 6x10"CeO2 5.411 0.12 8.22 6x 10-9 Y203 10.604 -1.89 6.22 6x 10-1' YSZ 5.139 -5.03 3.07 2x 10"8


TiN 4.242 10.43 18.49 20-30 LaMnO3 3.880 1.60 9.70 8x 10-15 Ni 3.524 5-6 Cu 3.615 1.5-2











Material Structure Lattice Constant Resistivity Applications for Growth Conductive Buffer Layer Method



rhombohedral [146] a = 5.461A, a = 60.41' - 3x 104 Q.cm @300 K YBCO / LNO / Ni tape [123]
pseudo-cubic [147] a - 3.83A - 7x 10- f-cm @77 K YBCO / SRO / LNO / Ni tape [83] rf
LaNiO3 T: 91 K magnetron J,: 1.3x 106 A/cM2 @ 0T sputtering


distorted orthorhombic [129] a = 5.573A, b = 5.538A, c = 7.586A -2x 104 .cm @300 K [125] YBCO / SRO / LAO [127]
pseudo-cubic a= 3.93A - 5x 10-5 .cm @77 K Tc: 91 K
SrRuO3 Jc: 3.0x 106 A/cm2 @ OT PLD "



rhombohedral or a = 3.88A [78] _ I X 10-3 .cm @300 K [149] YBCO / LSMO / Ni tape [133]
cubic [148] - 2x 104 Q-cm @77 K YBCO / LSMO / Ni / Cu tape [86] rf
Lao.7Sro.3MnO3 T: 91 K magnetron 4: 2.3x106 A/cm2 @ OT sputtering

orthorhombic (x -0.3) [140] ifx = 0.5 YBCO / LSTO / Ir / Cu tape
cubic (x > 0.7) - 1 x 10-4 .cm @300 K [2.4.70] T, : 86 K
Laj_,Sr.TiO3 j,: 1.0X106 A/cm2 @ 0T PLD



distorted orthorhombic [132] a = 3.835A - 5x 10-5 n-cm @300 K [143]
pseudo-cubic - 2.5 x 10-5 0-cm @77 K La0 sSro sCo03



distorted rhombohedral [150] a = 5.431 A, -60.5 1

LaCuO3



tetragonal (x -0.2) [145] a = 10.867A, c = 3.857A _ I x 10-3 0-cm @300 K [145]

Lai.0SrCuO20











N(E)


Eg



JE EF


(a)


1.0


0.8

U O 0.6 L&. I
0.4 LIJ


0.2


0.2 0-4 0.6 0.8 1.0
T/Tc


(b)


Figure 2-1. (a) Schematic drawing of energy gap in superconducting material, (b)
theenergy gap dependence on temperature with theoretical value versus
experimental results for some elemental metals.


0 Tantalumn


-Theo reti cal








After Blatt, Modern Physic











lt


Normal state


T T


H


Figure 2-2. Schematic drawing of the critical space within which the superconductivity
remains (T-H-I diagram).


Hc2(0)




H



HeI(O)


T Tc


Figure 2-3. Phase diagram of type II superconductor with temperature and magnetic field
dependency.












Cu02 Cu02 La Yk
charge
La reservoir


Cu02 Cu02

La charge reservoir
La C0

Cu2U02


Figure 2-4. Schematic drawing of La2Cu04.











CuO Ba-O

Yttrium CU02-, - .
'Barium YCopper 0 Oxygen



Ba-O CuO





Figure 2-5. Schematic drawing of YBa2Cu3O7-. unit cell combined with oxygen deficient
perovskite structure.




















40j


Figure 2-6. Schematic drawing of YBa2Cu306 compound which has different CuO chains
with YBa2Cu307.


4P Yttriumn

t Barium 6 Copper 0 Oxygen









BiO 0z

BiO

SrO~ CuO

SrO4 BiO.


4zEzz~


BiO 0 Bi e Sr 0 Ca


eCu oO


(a)


(b)


(c)


Figure 2-7. Schematic drawing of(a) Bi-Sr-Cu-O, (b) Bi-Sr-Ca-Cu-O (double Cu-O
sheets), and (c) Bi-Sr-Ca-Cu-O (triple Cu-O sheets).








Ni~lll)Ni ti111)



0 "0 0 0 0 00 000000 ROLLING ANNEALIN(;


Ni (111)

0


Ni (III)



0

11170


KLUU L/z1


Oxide ( 11)

0D O


BUFFER LAYER DEPOSTION


RABiTS


Figure 2-8. Conceptual drawing of rolling assisted biaxially textured substrate (RABiTS)
process based on Ni substrate. The pattern inside the circle designates the
texture alignment of the substrate.





41




1800

SAO= 5.4555'


.2A
L_k
CU



.4-aA
C






15 20 25 30 35

0 (deg)



Figure 2-9. The X-ray diffraction (9-scan (rocking curve) of (002) Ni which indicates the
substrate's out-of-plane alignent. The full-width-half-maximum (FWHM) of
the curve is 5.4555'.





42




10000
A

A+ = 7.7762'
A






.








-150 -100 -50 0 50 100 150 S(deg)



Figure 2-10. The x-ray diffraction f-scan of(I 11) Ni which indicates the substrate's inplane alignment. The average full-width-half-maximum (FWHM) of the 4
peaks is 7.7762�.







43














32.77 N \1382 5.83 2.46 1.04 0.44


0.08 0.03












Figure 2-11. The log-scale (111) Ni pole figure of the Ni substrate formed by RABiTS

process. The cube fraction of the substrate is 98.2694%.












70000




..

(U


1)


35 40 45
20 (deg)

(a)


70000


50 55


35 40 45 50 55
20 (deg)


(b)


Figure 2-12. The X-ray diffraction 0-20 scan graphs of textured Cu tapes formed by
RABiTS process (a) before annealing and (b) after annealing at 800'C in
vacuum for 2 hours.


0
0
0


Ha - ... . .. .


IJ


























34,96 14.74 262

LII

0.47 020 1.003













(a)















5135 4 S 2174 917



163 060 029



V 005












(b)




Figure 2-13. The log-scale pole figures of(1 11) Cu substrates (a) before annealing and

(b) after annealing at 800'C in vacuum for two hours.













( YBCO HTS layer -1 pm

Ce02 cap layer -* 0.75 pm YSZ barrier layer -+ 0.75 pm
_ _ _ __ Y203 seed layer -+ 0.75 gm

Figure 2-14. Schematic vertical structure drawing of conventional 2G wire based on
YBCO HTS.


Cope stailze a~lr- Solderw
ile


Figure 2-15. SEM cross-section of AMSC's 2G wire.










30.Op

25.Op

E 20.Op

P5.0 Ni-W (3 at.%) tape

"1 O.OpJ


I 0.0



50 100 150 200 250 300
Temperature (K)


Figure 2-16. Resistivity comparison of Cu tape and Ni-W (3 at.%) tape as a function of
the temperature.











200

Ni tape .... Cu tape j 150



100



. 50



0
0 200 400 600 800 1000

Current/tape-width (A/cm)



Figure 2-17. Simulated graph of cap layer thickness as a function of current per tape
width.











100
Ni tape
-------- Cu tape
80


360


~40


20


0
0 200 400 600 800 1000

Current/tape-width (A/cm)



Figure 2-18. Simulated graph of engineering current density as a function of current per
tape width.









lxlO&


lxlO1 CuO
.I CU ....

ixi03
-lxlO.37 j xl0-5

1 x10-71/C
1 x10.9
UC lxlO-13 [
105 [
200 400 600 800 1000 1200
Temperature (IC)


Figure 2-19. Thermodynamic stability curve of Cu in terms of temperature and oxygen
partial pressure.









Heating Block


Substrate
Material













to Vacuum iIl Pump


Ar


/ Magnetic
Field


/
Target


Gas Inlet Chamber Wall


Magnet


High Voltage


Figure 2-20. Schematic drawing of magnetron sputtering system.


N


r






52


Substrate
Material


Electron Gun Shield





to Vacuum
Pump


Charged Material Crucible



/


/
Filament


T


,-Electron Beam





- Chamber Wall


M
Magnet


Figure 2-21. Schematic drawing of electron beam evaporation system.








Scanning Mirror


Pulsed
-Laser
Beam

Ablation Plume





m

Substrate
Heater


to
Vacuum Pump


Figure 2-22. Schematic drawing of pulsed laser deposition system.


Vacuum






























Figure 2-23. Schematic drawing of typical perovskite crystal structure.


A insulator


U'

Figure 2-24. Schematic diagram of Zaanen, Sawatzky and Allen (ZSA) framework.


oxygen states


states


A

U' low U' metal













CHAPTER 3
GROWING EPITAXIAL CU2MG AS AN OXIDATION BARRIER FOR HIGH
TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS


3.1 Introduction

High-temperature Superconducting (HTS) biaxially-textured coated conductors hold significant promise for the development of a superconducting wire technology functional at the liquid nitrogen temperatures (64-77 K). To date, this technology utilizes epitaxial YBa2Cu307 (YBCO) coatings deposited on biaxially-textured Ni or Ni-based alloy substrates [35][39][41][151]. These tape substrates are fabricated by thermomechanical deformation of the metals. While HTS wires based on the biaxially textured Ni substrates offer the promise of substantial energy savings for applications in the power sector, the actual impact of HTS wire will be determined by several factors, including the reliability of these conductors against catastrophic failure, as well as the potential limitations introduced via the use of ferromagnetic substrates. One approach that addresses both substrate ferromagnetism and supercurrent shunting is to use high conductivity Cu tapes as the base metal substrate. For use as a substrate for HTS coated conductors, the oxidation of the Cu substrate must be addressed. A significant issue with Cu is the oxidation of the metal substrate due to the following buffer oxide and YBCO depositions at the elevated temperature with high pressure of oxygen gas. The copper oxide (Cu2O, CuO) growth occurs at the oxide-gas interface and the rate-determining step of oxidation is the diffusion of cation vacancies [152]. In general, the oxidation of Cu proceeds rapidly and at minimal oxygen partial pressure. At a temperature of 700'C,








Cu20 formation is favored for pure Cu exposed to oxygen pressures as low as 10-8 Torr. A typical RABiTS is composed of several buffer layers such as CeO2, Y203 and yttriastabilized zirconia (YSZ). At the temperatures of YBCO deposition, these oxides cannot block oxygen diffusion causing the oxidation of Cu substrate. Copper oxide is not an effective passivation layer to further oxidation or scaling.

In metallurgical applications, enhancement of Cu oxidation resistance has been investigated by several approaches, including alloying or implanting elements such as Mg, Cr and Al that are known to be oxygen getters [87]-[92]. Among these, the Mg-Cu and Al-Cu systems are perhaps the most attractive due to the limited solubility of Cr in Cu [93]. The use of Mg-doped Cu films as an oxygen diffusion barrier is being investigated for Cu metallization in Si integrated circuit technologies [941. Unfortunately, the resistivity of Cu increases rapidly with dopant concentration, deterring the use of an ally as the substrate material if shunting functionality is desired. However, Mg-Cu or AlCu alloy thin film on the Cu substrate may prove suitable since the effective electrical transport path length would be limited to a thin surface layer. The phase diagram for the Cu-Mg binary alloy system indicates the presence of intermetallic Cu2Mg at the range of 15-18 weight % Mg up to 819'C [153]. The crystal structure of Cu2Mg is cubic fcc with the lattice constant a = 7.064A [154]. In addition the room temperature resistivity of Cu2Mg is known to be 5-6 piLcm which is compatible with that of Cu (1.67 jncm) [155]. One can therefore consider two variants of (Mg,Cu) buffer layers for Cu substrates, namely Mg-doped fcc Cu and the cubic Cu2Mg intermetallic. In order to assess the potential applicability of this approach to oxidation resistant buffers for








RABiTS-based conductors, we have investigated the epitaxial growth of (Cu,Mg) films on a (001) Cu surface.

3.2 Experiments

For these experiments, epitaxial (001) Cu films on (001) MgO single crystals were used as the substrate materials. The Cu films were grown using sputter deposition at a substrate temperature of 400�C with 1.0x 10-2 Torr of Ar gas. The thickness of the epitaxial Cu film was 180nm. Due to the high vapor pressure of Mg (over 2.0x 10-3 Torr at 4000C), in situ growth of epitaxial (Cu,Mg) films is not possible. The approach used to achieve epitaxy of a (Cu,Mg) film was to form Cu/Mg multilayer precursor films that would be subsequently annealed to form either Mg-doped fcc Cu or intermetallic Cu2Mg. Sputter deposition was used to deposit Mg and Cu multilayers at room temperature with I.Ox 10-2 Torr of Ar gas. The precursor consists of an Mg/Cu multilayer stack with 5 each of 25 nm thick Mg and 25 nm thick Cu layers, which were grown at room temperature by sputter deposition. Figure 3-1 shows the vertical structure, which is used for this experiment. The precursor is then annealed in a flowing H2/Ar mixture at temperatures ranging from 400�C to 700'C. The development of (Cu,Mg) phases as the multilayer was investigated as a function of annealing temperature and duration.

3.3 Results and Discussion

At annealing temperature of 400�C, formation of the intermetallic Cu2Mg was observed. Figure 3-2 shows a 2-0 X-ray diffraction scan along the surface normal for a multilayer structure annealed at 400�C. The strongest peaks, other than that from the substrate and epitaxial Cu film, correspond to Cu2Mg. Both (111) and (001) oriented Cu2Mg grains are observed. Interestingly, the (001) oriented component of the








intermetallic phase was found to be epitaxial with respect to the C layer. Figure 3-3 shows the X-ray diffraction rocking curve for the (004) Cu2Mg peak, indicating a fullwidth half-maximum (FWHM) of AO = 2.0�, which is slightly larger than that for the Cu film (AO = 1.450). The in-plane orientation of the c-axis oriented Cu2Mg was investigated using four-circle x-ray diffraction. Figure 3-4 shows an in-plane (p-scan through the Cu2Mg (222). The four-fold symmetric peaks indicate that the grains are epitaxial with respect to the Cu film, possessing a cube-on-cube orientation. The FWIM of the in-plane peaks is 2.1'. The cubic Cu2Mg lattice parameter was measured to be 7.016 � 0.005 angstroms. At the upper end of the temperature range (600'C, 700*C), only the (001) Cu peaks are observed. This is consistent with the formation of an epitaxial Mg-doped fcc Cu alloy. Figure 3-5 is the 2-0 x-ray diffraction scan of the (CuMg) multilayer after annealing at 400, 500, 600, 700'C. In order to confirm the oxidation resistance of the structures possessing (Cu,Mg) alloy films, CeO2 was deposited at elevated temperatures on Ni / (CuMg) / Cu / MgO substrates and compared to CeO2 films on Ni / Cu / MgO. The CeO2 films were deposited by PLD at 750*C, and included a thin nucleation layer deposition in vacuum, followed by CeO2 deposition at 2x 10-4 Torr of oxygen. In case of the CeO2 film on Ni / Cu / MgO, significant surface roughness due to the metal oxidation is observed. This is shown in figure 3-6. In contrast, no surface roughness is observed in the SEM image for the CeO2 / Ni / (CuMg) / Cu / MgO structure.

3.4 Conclusions

The growth of epitaxial Cu2Mg as an oxidation barrier for high temperature superconducting coated conductors was investigated. Epitaxy of (004) Cu2Mg intermetallic phase was achieved on (002) Cu film. In-plane (p-scan through the Cu2Mg





59

(222) and the x-ray diffraction rocking curve for the (004) Cu2Mg peak indicates that the intermetallic phase is well oriented on (002) Cu surface.





60


Cu 25nm
25nm Mg
Cu 25nm
25nm Mg Cu 25nm
25nm Mg Cu 25nm
25ni Mg Cu 25nm
25nm Mg

Cu l8Onm


Figure 3-1. The vertical structure which is used for this experiment.







106
MgO (002)
Cu (002) 105
...




102


1120 30 40 50 60 70 20 (deg)


Figure 3-2. The X-ray diffraction 0-20 scan along the surface normal for a multilayer
structure annealed at 4000C.









1600 1400 1200 1000


O-scan through
T . . . . .._. . .


AO= 2.0'


800 600 400


200L
0
20


22 24 26 28 30 32


0 (deg)
Figure 3-3. The X-ray diffraction rocking curve for the (004) Cu2Mg peak, indicating a
full-width half-maximum (FWHM) of AO = 2.00.


Ci
4
C)
a) 4-


Cu 2Mg (004)






)-scan through the Cu2 Mg (222)

Ab = 2.10 FWHM


~J~4K~


-150-100 -50


0 50 100 150


I (deg)


Figure 3-4. The X-ray diffraction +-scan through the Cu2Mg (222).


.c


103


102










4000 ...I.. 1 1 '.
.()MgO (002)-m,- -.,Cu (002)


3000
..


S2000
(b)
1 w
00
MINE000S CU2Mg (004)"
(a) 10



5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 20 (deg)


Figure 3-5. The X-ray diffraction 0-20 scan of the (Cu,Mg) multilayer after annealing at
400, 500, 600, 7000C.

























Figure 3-6. The SEM picture of (a) CeO2 film on Ni / Cu / MgO (b) CeO2 / Ni / (Cu,Mg) /
Cu / MgO structure.













CHAPTER 4
(LA,SR)T103 AS A CONDUCTING BUFFER LAYER FOR HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS

4.1 Introduction
Among the potential application of high-temperature superconductivity (HTS) are the power and utility sectors that will require high-current wires [8]. In the RABiTS (rolling assisted bi-axially textured substrate) approach, HTS coated conductors are made by deposition of buffer layers and superconducting oxide on a crystalline-textured metal tape. This approach is conducive to the development of electrically conductive buffer layers, which would provide a current shunt to the metal tape for improved protection. There are key issues for a conductive buffer layer of HTS coated conductors. First, it must be reasonably well lattice matched to both the metal substrate and the superconducting film, thus enabling epitaxy. Second, the interaction between the buffer layer and the metal substrate must be such as to minimize formation of any native interfacial oxide that would serve as an insulating barrier to shunted current flow. Although buffer layers of the conductive oxides La0,7Sr0.3MnO3, SrRuO3 and LaNiO3 on textured Ni tape have been reported [83,133], to date most buffer layers investigated for HTS conductors have been insulating, such as CeO2 [156-158], Y203-ZrO2 [68,151], [159,160], ZrO2 [161], SrTiO3 [162] and NiO [163], [164]. In the latter situation, protection must be accomplished by the addition of an adjacent normal metal layer that is sufficiently conductive and thick to accommodate the current flow without unmitigated growth of an unstable hot zone.








An alternative candidate material system that may satisfy the criteria for conductive buffers mentioned above is (LaSr)TiO3. LaTiOa+x is an interesting defect perovskite system, with transport properties varying from insulating to metallic based on oxygen stoichiometry [138,140,165]. La and Ti have a relatively high affinity for oxygen. This suggests that the driving force for native oxide formation at the interface between LaTiO3 and either Ni or Cu should be reasonably low. In previous work, we have demonstrated epitaxial growth of LaTiO3 on (001) Ni using pulsed laser deposition [166]. The results indicate that epitaxy can be achieved, although the stability of the LaTiO3/Ni structure is limited to reducing conditions due to the phase transitions that occur with increased oxygen content. With an extreme sensitivity to oxygen content, LaTiO3+x is not particularly attractive as a conductive buffer layer. One possible approach to maintaining metallic conductivity in an oxidized state is through cation doping. This could maintain carrier density as well as reduce oxygen diffusivity. The most likely dopant candidate is Sr. Doping with divalent element increases the Ti 3 / Ti4 ratio and can make the compound less sensitive to ambient oxygen pressure during YBCO deposition. Electrical conductivity is a function of doping constant x in the LalSrxTiO3 compound, and the resistivity continuously decreases with higher cation doping [140]. Previous study on oxygen dependency showed the room temperature resistivity of La0.5Sr0.5TiO3 was remained below 1.0xl0-3 -cm in 10-4 - 10-2 Torr of oxygen pressure range [141].

In this chapter, the fundamental properties of (LaSr)TiO3 films on single crystal SrTiO3 substrates have been investigated and compared with the LaTiO3 films. The HTS films such as YBa2Cu307 grown on (LaSr)TiO3 buffer layers on single crystal SrTiO3








substrates have been also investigated in order to evaluate the compatibility of thin film layers.

4.2 Experiments

The growth of (LaSr)TiO3 and LaTiO3 films on (001) oriented single crystal

SrTiO3 substrate, 0.2-0.5gm thickness, was performed by pulsed laser deposition (PLD) in vacuum at 750'C, at an energy density of 2 J/cm2 and repetition rate of 10Hz. A KrF (248nm) excimer laser was used as the ablation source. The oxygen sensitivities of (La,Sr)TiO3 and LaTiO3 films have been investigated by changing the oxygen partial pressure from 3.x 10_5 Torr to 4.Ox 10-4 Torr. The high temperature superconducting (HTS) YBa2Cu307 (YBCO) layer was also grown by PLD under conditions of 1.Ox 10Torr of oxygen at 780�C, energy density of 2 J/cm2, 10Hz rate, with a thickness of

0.2Wn. After completing YBCO deposition, the PLD chamber was cooled down to 500'C at a rate of 280C/min and the oxygen pressure was increased to 400 Torr. After 20 minutes, the chamber was cooled down to room temperature rapidly.

The x-ray diffraction of 0-20 scan, o and k-scan were used in order to observe the thin film crystallinity, in-plane and out-of-plane alignments of each film layers. The in situ reflection high-energy electron diffraction (RHEED) was used to monitor epitaxial film growth. The film thickness was measured by step profilometer which converts the physical step distance into the electrical signals.

A standard four-point probe technique was used to evaluate the electrical

properties, including temperature-dependent resistivity (R-T) of not only the buffer layers but also the superconducting film, the superconducting transition temperature (T) and the critical current density (J,). The R-T measurements were done in the cryogenic cooling








system which was operated in the liquid helium of 4 K. The critical current density (J) as a function of magnetic field was evaluated by magnetization method in liquid helium. The widths of the samples were in the range of 0.28 -0.53cm and the distance between the voltage tips was in the range of 0.3-0.4cm.

4.3 Results and Discussion

Figure 4-1 shows the x-ray diffraction 0-20 scan of(LaSr)TiO3 film deposited on (001) oriented SrTiO3 single crystal substrate. The film was grown by pulsed laser deposition (PLD) method at 750'C in vacuum. The y-axis which indicates the intensity of figure 4-1 is designated by log scale in order to separate the film and substrate peaks clearly. The x-ray 0-20 scan of (LaSr)TiO3 film shows (001), (002), (003) peaks at 22.280, 45.50', and 70.93' respectively. The cubic lattice constant of the film is 3.98A calculated from the Bragg's law. Figure 4-2 is the in situ monitored reflection highenergy electron diffraction (RHEED) pattern of (LaSr)TiO3 film grown at 750�C in vacuum. According to the x-ray 0-20 scan and RHEED pattern, the (LaSr)TiO3 film is epitaxially deposited on SrTiO3 single crystal substrate by PLD method at 750'C in vacuum. In order to examine the oxygen sensitivity, the (LaSr)TiO3 films were grown in oxygen ambient with partial pressure of 3.0x 10-5 Torr and 4.0x 10-4 Torr. Figure 4-3 shows the x-ray diffraction 0-20 scan of (LaSr)TiO3 films deposited on SrTiO3 single crystal substrates by PLD method with different ambient conditions. Only the (002) peaks of each sample are designated in this figure to make clear comparison. The dashed lines indicate the standard SrTiO3 (002) and LaTiO3 (002) peak positions at 46.48', 46.03' respectively. The films grown under the oxygen partial pressure show the (002) peaks at almost the same position with LaTiO3 (002) at 46.03'. This indicates that the








oxygen deficiency in the film is compensated by ambient oxygen pressure leading to the stoicheometric LaTiO3 film. Figure 4-4 shows the x-ray diffraction 0-20 scan of LaTiO3 film deposited on (001) oriented SrTiO3 single crystal substrate. The film was grown by pulsed laser deposition (PLD) method at 750'C in vacuum. The x-ray 0-20 scan of LaTiO3 film shows (001), (002), (003) peaks at 22.26', 45.47', and 70.890 respectively. The cubic lattice constant of the film is calculated by 3.98A. In order to compare the oxygen sensitivity of LaTiO3 with (La,Sr)TiO3 film, same oxygen partial pressure conditions such as 3.Ox 10-5 Torr, 4.Ox 104 Torr were applied during LaTiO3 deposition. In the figure 4-5 (a) and figure 4-6 (a), the x-ray peaks were taken with aligning the sample to the (001) oriented SrTiO3 substrate. There are no clear peaks which designate the film existence in both cases. In contrast with LaTiO3, La2Ti207 which is a fully oxidized phase shows monoclinic layered structure. The crystallographic angle between (-210) La2Ti207 and (001) SrTiO3 is known to be 4.52' [4.13]. The 0-20 scan graphs in the figure 4-5 (b) and figure 4-6 (b) were taken by tilting the substrate around 4' off axis. According to the figure 4-5 and figure 4-6, LaTiO3 films which are growing under the oxygen ambient can be turned into the fully oxidized La2Ti207 phase. Figure 4-7 shows the x-ray diffraction k-scan of (-420) peak for the La2Ti207 layer grown under the oxygen pressure of 3.0x10-5 Torr. The fourfold symmetry of the peak indicates that the film is inplane aligned.

In addition to the structure, the transport behavior of epitaxial (LaSr)TiO3 and LaTiO3 films grown on SrTiO3 is indicated in figure 4-8 as a function of the oxygen pressure during growth. Figure 4-8 (a) designates the resistivities measured at 300 K and figure 4-8 (b) shows the resistivities measured at 77 K. In case of the vacuum growth








condition, both (LaSr)TiO3 and LaTiO3 films show low and nearly identical values, with a resistivity on the order of 10-5 Q cm at 300 K. As oxygen pressure is moderately increased up to 4.Ox 104 Torr, the resistivity of LaTiO3 film increases by more than two orders of magnitude to -'0.2 Q cm at 300 K. This increase in the resistivity with oxygen pressure during growth is related to the structural transition from LaTiO3 to La2Ti207 phase as indicated in the X-ray diffraction pattern of figure 4-5 and figure 4-6. In contrast, the resistivity of (LaSr)TiO3 films is relatively insensitive to oxygen pressure making it attractive as a conductive buffer for coated conductor applications. Figure 4-9 and 4-10 show the resistivity curves of(La, Sr)TiO3 and LaTiO3 films on SrTiO3 substrates under various oxygen pressures as a function of measuring temperature. All of the R-T curves indicate that the films show metallic behaviors.

Based on the fundamental results of(LaSr)TiO3 films on single crystal SrTiO3 substrate, high temperature superconducting YBa2Cu307 layer was deposited on (La,Sr)TiO3. In this case, the (LaSr)TiO3 layer played a role of buffer layer similar to the real coated conductor application such as RABiTS. Figure 4-11 is the X-ray diffraction 020 scan of YBa2Cu307 film grown on (LaSr)TiO3 buffer layer on single crystal SrTiO3 substrate. The (001) peaks of YBa2Cu307 film are well defined. Figure 4-12 shows the Xray diffraction o-scan of (002) peak for (LaSr)TiO3 buffer layer grown on SrTiO3 substrate and (006) peak for YBa2Cu307 HTS film deposited on (LaSr)TiO3 buffer layer. The full width half maximum (FWHM) values of each layer are 0.970 and 1.58', respectively. Figure 4-13 shows the X-ray diffraction 4-scan of (112) peak for (La,Sr)TiO3 buffer layer grown on SrTiO3 substrate and (012) peak for YBa2Cu307 HTS film deposited on (LaSr)TiO3 buffer layer. The average full width half maximum








(FWHM) values of the fourfold symmetric peaks for each layer are 0.710 and 1.48', respectively. According to the 0)-scan and 4-scan, the YBa2Cu307 film is in-plane and out-of-plane aligned to the (LaSr)TiO3 buffer layer. Figure 4-14 shows the four-probe resistivity measurement for the YBa2Cu307 / (LaSr)TiO3 / SrTiO3 structure. The sample has the superconducting transition temperature (T) of 91 K indicating cationcontamination-free YBa2Cu307. Figure 4-15 is the critical current density (J) as a function of the magnetic field of YBa2Cu307 / (LaSr)TiO3 / SrTiO3 structure. The J, values were evaluated by magnetization method in liquid helium. The Jc can be calculated by the equation of J. = (15 xAM) / r, here AM is the magnetization difference and r is the radius of the sample. In this graph, the zero-field transport Jc value is

2.18x 106 A/cm2. The interesting aspect of the conductive buffer layer in the coated conductor application is whether its conductivity remains after HTS film deposition or not. Normally, the HTS film such as YBa2Cu307 is grown at high temperature in high oxygen pressure. The underlying buffer layer should be exposed to this extreme oxidizing condition. The resistivity of (LaSr)TiO3 film after annealing at YBa2Cu307 deposition condition (at 780'C, Po2 1.oxlo- Torr). Figure 4-16 shows the resistivity curve of (La,Sr)TiO3 film grown on SrTiO3 substrate after annealing at YBa2Cu307 deposition condition. The resistivity values at 300 K and 77 K were about 1.0 Q cm and 0.09 Q cm. The resistivity values of (LaSr)TiO3 film without annealing were 3.5x 10-5 Q cm and

1.5x 10-6 Q cm at 300 K and 77 K, respectively. Although the absolute values were increased by roughly 104 orders of magnitude, the resistivity curve indicates that the (La,Sr)TiO3 film remains metallic until the liquid nitrogen temperature of 77 K. The resistivity results of post annealed sample at YBa2Cu307 deposition condition show that








the (LaSr)TiO3 layer can be a candidate for the conductive buffer layer in the coated conductor applications.

4.4 Conclusions

The perovskite (LaSr)TiO3 was investigated as a possible conducting oxide buffer layer for high temperature superconducting coated conductors. In order to observe the oxygen sensitivity, thin (LaSr)TiO3 films were epitaxially deposited by PLD on single crystal SrTiO3 substrates at various oxygen partial pressures and compared with LaTiO3 films at the same growth conditions. The room temperature resistivity of LaTiO3 increases rapidly as the oxygen pressure increases by more than two orders of magnitude. In contrast, the resistivity of (LaSr)TiO3 films is relatively insensitive to oxygen pressure making it attractive as a conductive buffer for coated conductor applications.

The high temperature superconducting layer such as YBa2Cu307 was grown

epitaxially on (LaSr)TiO3 buffer layer on SrTiO3 substrate with excellent in-plane and out-of-plane alignment. The superconducting transition temperature (Tc) of YBa2Cu307 / (La,Sr)TiO3 / SrTiO3 structure was 91 K and the critical current density (Jc) of this structure was 2.18x 106 A/cm2 at 0 magnetic field. The resistivity results of post annealed sample at YBa2Cu307 deposition condition indicates that the (LaSr)TiO3 layer can be a candidate for the conductive buffer layer in the coated conductor applications.












106S
U" jo6 0
'105




103

102
opU)
U)


101

100
20 40 60 80 20t (degrees)


Figure 4-1. The X-ray diffraction 0-20 scan of (LaSr)TiO3 film grown on SrTiO3 single
crystal substrate by pulsed laser deposition (PLD) method at 750C, in
vacuum.
































Figure 4-2. The in situ reflection high-energy electron diffraction (RHEED) pattern of
(La,Sr)TiO3 film deposited on SrTiO3 single crystal substrate by PLD method
at 750'C, in vacuum.RHEED is used to monitor the epitaxial film growth.












__106



105 in vac.
002 :3e-5 Torr
j' 4 02 : 4e-4 Torr
104 2



103 LTO (002)
STO (002)

102
44 46 48 20 (degrees)



Figure 4-3. The X-ray 0-20 scans of (La,Sr)TiO3 films deposited on SrTiO3 single crystal
substrates in different ambient conditions.










107

106



104
~103




, 102 .lol0
-0 -m L IR

I00




20 40 60 80

20 (degrees)


Figure 4-4. The X-ray diffraction 0-20 scan of LaTiO3 film grown on SrTiO3 single
crystal substrate by pulsed laser deposition (PLD) method at 750'C, in
vacuum.














a 04
- 0 104 O
_A 0






00 4 0 100 (b)





20 40 60 80

20 (degrees)



Figure 4-5. The two different X-ray 0-20 scans of LaTiO3 film grown on SrTiO3 single
crystal substrate in the oxygen pressure of 3.0 10-5 Torr, aligned to (a) the
substrate (001) plane and (b) the La2Ti207 (-210) plane, respectively.















104
_ POO

100


10-4





20 40 60 80 20 (degrees)


Figure 4-6. The two different X-ray 0-20 scans of LaTiO3 fillmgrown on SrMi3 single
crystal substrate in the oxygen pressure of 4.0� 10 4 Torr, aligned to (a) the
substrate (00 1) plane and (b) the La2Ti207 (-2 10) plane, respectively.













O 03N A0 =15.350
*





*N







101
0 90 180 270 360


0 (degrees)

Figure 4-7. The X-ray +-scan of (-420) peak for La2Ti207 layer grown on SrTiO3 single
crystal substrate in the oxygen pressure of 3.Ox 10-5 Torr.













LaTiO3








-0 (La,Sr)TiO 3


(b)


LaTiO3





S (L2,Sr)TiO3


in vac.


10, lxlO1 lxlO-4 I

02 Pressure (Torr)


Figure 4-8. The resistivities of (LaSr)TiO3 and LaTiO3 films on single crystal SrTiO3,
measured at (a) 300 K and (b) 77 K, as a function of the oxygen pressure.


r (a)


r


r





in vac.
..a |


105 10'

3


10
102 ~10,


101 1
















(a)












....e .. .... ... .


0 50 100 150 200

Temperature


Figure 4-9. The resistivity curves of (LaSr)TiO3 films on SrTiO3 substrates grown in (a)
vacuum, (b) 3.0x 10-5 Torr of oxygen, and (c) 4.Ox 10-4 Torr of oxygen, as a
function of temperature.


ltii


J3Wt - -.-.. . . . . . . -. -.-. . . . . .


300 2250 j200 150

100 ~50

0


250

(K)


300 350


I . . . . B . . . . | . . . . � . . . . �






83





60

50

~40 (a) 0 .30 . 20 10












16 000 ...... . ......... '.... .........
800 (b) 600W ~400 M 200

0






1680000





8, 0000





0
0 50 100 150 200 250 300 350 Temperature (K) Figure 4-10. The resistivity curves of LaTiO3 films on SrTiO3 substrates grown in (a)
vacuum, (b) 3.x10-5 Torr of oxygen, and (c) 4.Ox104 Torr of oxygen, as a
function of temperature.











8000

8 OO





- 0




00




5 10 15 20 25 30 35 40 45 50 55 60

20 (degrees)



Figure 4-11. The X-ray 0-20 scan of YBa2Cu307 deposited on (LaSr)TiO3 buffer layer
on SrTiO3 single crystal substrate.





85



10000


AO =0.97
O(a)








6W
Cu










6000.

AO= 1.58 � (b)













0
15 20 25 30

0 (degrees)


Figure 4-12. The X-ray o-scan of (a) (002) peak for (LaSr)TiO3 buffer layer on SrTiO3
single crystal substrate and (b) (006) peak for the YBa2Cu3O7 film deposited
on (LaSr)TiO3 buffer layer.




Full Text

PAGE 1

NUCLEATION AND EPITAXY OF CONDUCTIVE BUFFERS ON (001) Cu FOR COATED HIGH-TEMPERATURE SUPERCONDUCTING CONDUCTORS By KYUNGHOON KIM 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 2005

PAGE 2

Copyright 2005 by Kyunghoon Kim

PAGE 3

To my family for their love and encouragement

PAGE 4

ACKNOWLEDGMENTS Above all, I would like to express my sincere appreciation to my advisor, Professor David P. Norton, for his excellent guidance and invaluable help. I cannot forget his warm support and the five years that I worked for Dr. Norton was the precious time in my life. My appreciation also goes to my committee members, Professor Cammy R. Abernathy, Professor Rajiv Singh, Professor Wolfgang Sigmund, and Professor Andrew Rinzler. Since I started collaboration research in Oak Ridge National Laboratory, I received a lot of support from the staff members and post doctors. I would like to give my deep appreciation to Dr. David K. Christen, who is my supervisor and a group leader of the superconductivity group. Dr. Christen gave made me feel comfortable staying in ORNL as a graduate student. I wish to give my special thanks to the members of Condensed Matter Sciences Division, Metals and Ceramics Division, and Chemical Science Division: Dr. Claudia Cantoni, who helped me use the PLD system and shared a lot of time to discuss technical issues; Dr. Tolga Aytug, who helped me use the sputtering system and gave me great advice; Dr. Albert A. Gapud, who helped me use the transport measuring system; Professor James R. Thompson, who helped me use the SQUID system and gave me lots of technical information; Yifei Zhang, who helped me use the SQUID system; Dr. M. Paranthaman, who gave me valuable help and encouraged me; Dr. Amit Goyal, who supported metal substrates; Patrick M. Martin, who helped me use electrical measuring instruments and annealing system; Dr. Ho Nyung Lee, who helped me get a IV

PAGE 5

lot of physical results and shared precious time to discuss technical issues; Dr. Sukill Kang, who helped me use various instruments and gave me warm support; and Dr. Daeho Kim and Dr. Mina Yoon who taught me lots of fundamental physics. 1 also want to thank all the members of Dr. NortonÂ’s group. I would like to acknowledge the enormous help of my sister-in-law, Youri, and her family. They gave me a lot of joy and happiness. I also express my deep appreciation to my elder sister, Hyeonjoo, who always encouraged me and prayed for my family. I would like to give my special thanks to my parents-in-law for their great affection and concern. Especially, it is my great honor to dedicate this work to my father-in-law who passed away last February. I am also deeply grateful to my parents, with all my respect, for their endless love, patience and prayers for me and my family. Finally, my great gratitude goes to my lovely wife, Youmi Kang, and my son, Youngwook Kim. I sincerely appreciate my wifeÂ’s encouragement, support and love. Without her love and support, I could not do all of this. v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS iv LIST OF TABLES viii LIST OF FIGURES ix ABSTRACT xiv CHAPTERS 1 INTRODUCTION 1 2 LITERATURE REVIEW 4 2.1 High Temperature Superconductors 4 2.2 Superconducting Power Applications using HTS Wire 9 2.2.1 Biaxially Textured Metal Substrates 1 1 2.2.2 Buffer Layers 13 2.3 Review of Conducting Oxide Buffers for Cu based YBCO Coated Conductors. 16 2.3.1 Introduction 16 2.3.2 Preventing Methods of Cu Oxidation 20 2.3.3 Thin Film Techniques for Oxides Growth 22 2.3.4 Overview of Conductive Oxides for HTS Coated Conductors 25 3 GROWING EPITAXIAL Cu 2 Mg AS AN OXIDATION BARRIER FOR HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 55 3.1 Introduction 55 3.2 Experiments 57 3.3 Results and Discussion 57 3.4 Conclusions 58 4 (La,Sr)Ti0 3 AS A CONDUCTING BUFFER LAYER FOR HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 66 4.1 Introduction 66 4.2 Experiments 68 4.3 Results and Discussion 69 vi

PAGE 7

4.4 Conclusions 73 5 EPITAXIAL (La,Sr)Ti0 3 AS A CONDUCTIVE BUFFER FOR Ni-W BASED HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 90 5.1 Introduction 90 5.2 Experiments 92 5.3 Results and Discussion 93 5.4 Conclusions 96 6 EPITAXIAL (La,Sr)Ti0 3 AS A CONDUCTIVE BUFFER FOR Cu BASED HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 11 1 6. 1 Introduction 1 1 1 6.2 Experiments 113 6.3 Results and Discussion 1 14 6.4 Conclusions 117 7 SUMMARY 131 LIST OF REFERENCES 134 BIOGRAPHICAL SKETCH 145 vii

PAGE 8

LIST OF TABLES Table page 2-1. The critical temperature values in absolute temperature unit reported for YBa 2 (Cui. x M x ) 307_5 systems 3 1 2-2. The list of buffer layer materials tested with RABiTS applications 32 2-3. The overall electrical and structural properties of perovskite oxide compounds mentioned in this section 33

PAGE 9

LIST OF FIGURES Figure Eige 2-1 . (a) Schematic drawing of energy gap in superconducting material, (b) theenergy gap dependence on temperature with theoretical value versus experimental results for some elemental metals 34 2-2. Schematic drawing of the critical space within which the superconductivity remains (T-H-I diagram) 35 2-3. Phase diagram of type II superconductor with temperature and magnetic field dependency 35 2-4. Schematic drawing of La 2 Cu 04 36 2-5. Schematic drawing of YBa 2 Cu 30 7 .6 unit cell combined with oxygen deficient perovskite structure 37 2-6. Schematic drawing of YBa 2 Cu 3 06 compound which has different CuO chains with YBa 2 Cu 3 0 7 38 2-7. Schematic drawing of (a) Bi-Sr-Cu-O, (b) Bi-Sr-Ca-Cu-O (double Cu-0 sheets), and (c) Bi-Sr-Ca-Cu-0 (triple Cu-0 sheets) 39 2-8. Conceptual drawing of rolling assisted biaxially textured substrate (RABiTS) process based on Ni substrate. The pattern inside the circle designates the texture alignment of the substrate 40 2-9. The X-ray diffraction co-scan (rocking curve) of (002) Ni which indicates the substrate’s out-of-plane alignment. The full-width-half-maximum (FWHM) of the curve is 5.4555° 41 2-10. The x-ray diffraction <|)-scan of (1 1 1) Ni which indicates the substrate’s in-plane alignment. The average full-width-half-maximum (FWHM) of the 4 peaks is 7.7762° 42 2-11. The log-scale (1 1 1) Ni pole figure of the Ni substrate formed by RABiTS process. The cube fraction of the substrate is 98.2694% 43 IX

PAGE 10

2-12. The X-ray diffraction 9-20 scan graphs of textured Cu tapes formed by RABiTS process (a) before annealing and (b) after annealing at 800°C in vacuum for 2 hours 44 2-13. The log-scale pole figures of (1 1 1) Cu substrates (a) before annealing and (b) after annealing at 800°C in vacuum for two hours 45 2-14. Schematic vertical structure drawing of conventional 2G wire based on YBCO HTS 46 2-15. SEM cross-section of AMSC's 2G wire 46 2-16. Resistivity comparison of Cu tape and Ni-W (3 at.%) tape as a function of the temperature 47 2-17. Simulated graph of cap layer thickness as a function of current per tape width 48 2-18. Simulated graph of engineering current density as a function of current per tape width 49 2-19. Thermodynamic stability curve of Cu in terms of temperature and oxygen partial pressure 50 2-20. Schematic drawing of magnetron sputtering system 51 2-21. Schematic drawing of electron beam evaporation system 52 2-22. Schematic drawing of pulsed laser deposition system 53 2-23. Schematic drawing of typical perovskite crystal structure 54 224. Schematic diagram ofZaanen, Sawatzky and Allen (ZSA) framework 54 31. The vertical structure which is used for this experiment 60 3-2. The X-ray diffraction 9-29 scan along the surface normal for a multilayer structure annealed at 400°C 61 3-3. The X-ray diffraction rocking curve for the (004) CmMg peak, indicating a fullwidth half-maximum (FWHM) of A9 = 2.0° 62 3-4. The X-ray diffraction (|)-scan through the CmMg (222) 63 3-5. The X-ray diffraction 9-20 scan of the (Cu.Mg) multilayer after annealing at 400, 500, 600, 700°C 64 3-6. The SEM picture of (a) CeCE film on Ni / Cu / MgO (b) CeCE/ Ni / (Cu.Mg) / Cu / MgO structure 65 x

PAGE 11

4-1. The X-ray diffraction 9-20 scan of (La,Sr)Ti 03 film grown on SrTiCE single crystal substrate by pulsed laser deposition (PLD) method at 750°C, in vacuum 74 4-2. The in situ reflection high-energy electron diffraction (RHEED) pattern of (La,Sr)Ti 03 film deposited on SrTiCE single crystal substrate by PLD method at 750°C, in vacuum. RHEED is used to monitor the epitaxial film growth 75 4-3. The X-ray 0-20 scans of (La.SrfTiCE films deposited on SrTiCE single crystal substrates in different ambient conditions 76 4-4. The X-ray diffraction 0-20 scan of LaTiCE film grown on SrTiO? single crystal substrate by pulsed laser deposition (PLD) method at 750°C, in vacuum 77 4-5. The two different X-ray 0-20 scans of LaTi0 3 film grown on SrTi03 single crystal substrate in the oxygen pressure of 3.0xl0‘ : ' Torr, aligned to (a) the substrate (001) plane and (b) the LaiTECE (-210) plane, respectively 78 4-6. The two different X-ray 0-20 scans of LaTiCE film grown on SrTiCE single crystal substrate in the oxygen pressure of 4.0x 1 O' 4 Torr. aligned to (a) the substrate (00 1 ) plane and (b) the La 2 Ti 207 (-210) plane, respectively 79 4-7. The X-ray 7 layer grown on SrTiCE single crystal substrate in the oxygen pressure of 3. Ox 10° Torr 80 4-8. The resistivities of (La.SrfTiCE and LaTiCE films on single crystal SrTiCE, measured at (a) 300 K and (b) 77 K, as a function of the oxygen pressure 8 1 4-9. The resistivity curves of (La,Sr)Ti 03 films on SrTiCE substrates grown in (a) vacuum, (b) 3. Ox 1 0° Torr of oxygen, and (c) 4. Ox 1 O' 4 Torr of oxygen, as a function of temperature 82 4-10. The resistivity curves of LaTiCE films on SrTiCE substrates grown in (a) vacuum, (b) 3.0x1 0° Torr of oxygen, and (c) 4.0x1 0‘ 4 Torr of oxygen, as a function of temperature 83 4-11. The X-ray 0-20 scan of YBa2Cu307 deposited on (La,Sr)Ti03 buffer layer on SrTi03 single crystal substrate 84 4-12. The X-ray co-scan of (a) (002) peak for (La,Sr)Ti 03 buffer layer on SrTiCE single crystal substrate and (b) (006) peak for the YBaiC^CE film deposited on (La.Sr)TiCE buffer layer 85 4-13. The X-ray <|>-scan of (a) (112) peak for (La,Sr)Ti 03 buffer layer on SrTiCE single crystal substrate and (b) (012) peak for the YBa 2 Cu 307 film deposited on (La.Sr)Ti 03 buffer layer 86 xi

PAGE 12

4-14. The resistivity versus temperature measurement for YBaiCuaOy film deposited on (La,Sr)Ti03 buffer layer on single crystal SrTiCE substrate 87 4-15. The critical current density (Jc) as a function of magnetic field of YBa2Cu3C>7 film deposited on (La,Sr)Ti03 buffer layer on single crystal SrTiCE substrate 88 416. The resistivity versus temp, graph of (La.Sr)TiC>3 film grown on single crystal SrTiCE substrate after annealing with the YBa2Cu3C>7 deposition condition 89 51. The X-ray diffraction 0-20 scan of (La,Sr)Ti03 films deposited directly on Ni-W tape by PLD at the temperature of (a) 700°C, (b) 750°C, and (c) 800°C 97 5-2. The energy dispersive X-ray spectroscopy (EDS) results of TiN films grown on Cu layers on single crystal SrTiCE substrates with the thicknesses of (a) 500A, (b) 1000A, (c) 2000A, and (d) 5000A. These curves were taken after annealing the samples at 740°C, in vacuum for 60 minutes 98 5-3. The atomic percent of Cu observed on the surface of TiN layer with the thickness range of 500-5000A. The square symbols designate the samples without annealing. The circles and triangles show the samples with annealing at 740°C, in vacuum for 30 minutes and 60 minutes, respectively 99 5-4. The X-ray diffraction 0-20 scan of (a) TiN layer deposited on textured Ni-W alloy tape, and (b) (La,Sr)Ti03 film deposited on TiN seed layer 100 5-5. The in situ reflection high-energy electron diffraction (RHEED) pattern of (a) TiN film deposited on textured Ni-W alloy tape, and (b) (La,Sr)TiC>3 film deposited on TiN seed layer 101 5-6. The X-ray diffraction co-scan of (a) Ni-W (002), (b) TiN (002), and (c) (La,Sr)Ti03 (004) planes 102 5-7. The X-ray diffraction ()>-scan of (a) Ni-W (1 1 1), (b) TiN (111), and (c) (La.Sr)Ti03 (112) planes 103 5-8. The X-ray diffraction 0-20 scan of high temperature superconducting YBa2Cu3C>7 film grown on (La,Sr)Ti03 buffer layer / TiN seed layer / Ni-W tape 104 5-9. The X-ray diffraction co-scan of (005) YBa2Cu307 which was grown on (La,Sr)Ti03 buffer layer / TiN seed layer / Ni-W tape 105 5-10. The X-ray diffraction c|)-scan of (103) YBa2Cu30 7 which was grown on (La,Sr)Ti03 buffer layer / TiN seed layer / Ni-W tape 106 5-11. The resistivity versus temperature graph for YBa2Cu3C>7 film grown on (La.SrfTiCE buffer layer / TiN seed layer / Ni-W tape 1 07 Xll

PAGE 13

108 5-12. The voltage versus current graph of YBa 2 Cu 30 7 film deposited on (La.Sr)Ti0 3 buffer layer / TiN seed layer / Ni-W tape 5-13. The SEM picture of YBa 2 Cu 30 7 film surface deposited on (La.Sr)Ti 03 buffer layer / TiN seed layer / Ni-W tape 109 514. The energy dispersive X-ray spectroscopy (EDS) results of YBa 2 Cu 30 7 film grown on (La,Sr)Ti0 3 / TiN / Ni-W tape 1 1 0 61. The surface SEM picture of the (La,Sr)Ti0 3 film grown on Cu tape after annealing at 780°C in oxygen partial pressure of 1 .Ox 1 O' 1 Torr for 7 minutes 1 1 8 6-2. The energy dispersive X-ray spectroscopy (EDS) graph of (La.Sr)Ti0 3 film grown on Cu tape after annealing at 780°C in oxygen 1.0x10"' Torr for 7 minutes 119 6-3 . 1 he X-ray diffraction 0-20 scan of (a) Ir layer deposited on textured Cu by PLD, and (b) (La.Sr)Ti0 3 film deposited on Ir layer 120 6-4. The in situ reflection high-energy electron diffraction (RHEED) pattern of Ir film deposited on textured Cu tape, and (b) (La.Sr)TiC >3 film deposited on Ir layer 121 6-5. The X-ray diffraction co-scan of (a) Cu (002), (b) Ir (002), and (c) (La,Sr)Ti0 3 (004) planes 122 6-6. The X-ray diffraction 3 film on SrTi0 3 substrate and (La,Sr)Ti0 3 film on Ir on SrTi0 3 substrate 125 6-9. The resistivity curves of Ir on SrTi0 3 single crystal with and without annealing at 780°C in oxygen 1.0x1 0‘ 4 Torr for 7 minutes 126 6-10. The X-ray diffraction 0-20 scan of YBa 2 Cu 3 0 7 layer deposited on (La,Sr)Ti0 3 / Ir multi buffer stack on textured Cu tape 127 6-11. The surface SEM picture of the YBa 2 Cu 3 0 7 film grown on (La,Sr)Ti0 3 / Ir buffer stack on Cu tape with the magnification of (a) xlOOO and (b) x5000 128 6-12. The energy dispersive X-ray spectroscopy (EDS) graph of (a) normal YBa 2 Cu 30 7 film surface, (b) defect region of YBa 2 Cu 30 7 film 129 6-13. The resistivity versus temperature measurement for YBa 2 Cu 30 7 film deposited on (La,Sr)Ti0 3 / Ir multi buffer stack on Cu tape 130 Xlll

PAGE 14

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 NUCLEATION AND EPITAXY OF CONDUCTIVE BUFFERS ON (001) Cu FOR COATED HIGH-TEMPERATURE SUPERCONDUCTING CONDUCTORS By Kyunghoon Kim August 2005 Chair: David P. Norton Major Department: Materials Science and Engineering In the 2 nd generation wire technology of high temperature superconducting coated conductors, highly conductive and nonmagnetic Cu substrate can improve the wire properties along with the conductive buffer layers, offering fully conductive wire architecture. This scheme requires two components, namely oxidation resistance for the Cu tape and conductive buffer layers. The growth of epitaxial Cu 2 Mg as an oxidation barrier was investigated. Epitaxy of (004) Cu 2 Mg intermetallic phase was achieved on (002) Cu film. An in-plane cp-scan through the Cu 2 Mg (222) and the x-ray diffraction rocking curve for the (004) Cu 2 Mg peak indicates that the intermetallic phase is well oriented on (002) Cu surface. The perovskite (La,Sr)Ti 03 was investigated as a possible conducting oxide buffer layer for high temperature superconducting coated conductors. YBa 2 Cu 3 C >7 was grown epitaxially on (La,Sr)Ti 03 buffer layer on SrTiC >3 substrate with excellent in-plane and out-of-plane alignment. The superconducting transition temperature (T c ) of YBa 2 Cu 3 C >7 / xiv

PAGE 15

(La,Sr)Ti 03 / SrTiCh structure was 91 K and the critical current density (J c ) of this structure was 2.18xl0 6 A/cm 2 at 0 magnetic field. The resistivity results of a post annealed sample at YBa 2 Cu 307 deposition condition indicates that the (La,Sr)TiC >3 layer can be a candidate for the conductive buffer layer in the coated conductor applications. The epitaxial film growth of (La,Sr)Ti 03 was examined on Ni-W metal alloy tape. The transition metal nitride such as TiN was deposited epitaxially on Ni-W tape by PLD and played an excellent role as a seed layer for (La,Sr)Ti 03 film growth on Ni-W tape. The YBa 2 Cu 3 C >7 film was deposited epitaxially on the (La,Sr)TiC >3 buffer layer with the TiN seed layer on Ni-W tape. The YBa 2 Cu 3 C >7 film grown on (La,Sr)TiC >3 / TiN / Ni-W tape has T c of 89 K and J c of 0.42x 1 0 6 A/cm 2 . The epitaxial film growth of (La,Sr)Ti 03 was examined on Cu tape as a possible conducting buffer layer for high temperature superconducting coated conductors. The noble metal, such as Ir, was deposited epitaxially on Cu tape by PLD for an oxygen diffusion barrier. The YBa 2 Cu 3 C >7 film was deposited epitaxially on the (La,Sr)TiC >3 and Ir buffer stack on Cu tape. The YBa 2 Cu 307 film grown on the (La,Sr)Ti 03 / Ir / Cu tape has a superconducting transition temperature of 90 K and a critical current density value of 1.0x10 A/cm . This shows that (La,Sr)TiC >3 is a possible candidate for the conductive buffer layer in the Cu based RABiTS applications. xv

PAGE 16

CHAPTER 1 INTRODUCTION The coated high-temperature superconducting (HTS) conductors consist of metal substrate, buffer layers and thin film superconducting oxide. The rolling-assisted biaxially textured substrate (RABiTS) process made it possible to offer long length HTS wire. The advantages of Cu tape with nonmagnetic, lower material cost (-20% of Ni), easy formation of a sharp cube texture and higher electrical and thermal conductivity (for Cu, k : 398 W/m-K and for Ni, k : 90 W/m-K) than Ni or Ni based alloy tapes can improve the YBa2Cu3C>7 based 2G wire properties along with the conductive buffer layers, which offers fully conductive wire architecture. Though the Cu based RABiTS process with the conductive buffer architecture has advantages, the Cu oxidation problem could be worse. In metallurgical applications, enhancement of Cu oxidation resistance has been investigated by several approaches. Among these, the use of Mg-doped Cu films as an oxygen diffusion barrier is being investigated with the intermetallic phase of Cu2Mg. There are key issues for a conductive buffer layer of HTS coated conductors. First, it must be reasonably well lattice matched to both the metal substrate and the superconducting film, thus enabling epitaxy. Second, the interaction between the buffer layer and the metal substrate must be such as to minimize formation of any native interfacial oxide that would serve as an insulating barrier to shunted current flow. An alternative candidate material system that may satisfy the criteria for conductive buffers mentioned above is (La,Sr)TiC>3. LaTi03 +x is an interesting defect perovskite system, with 1

PAGE 17

2 transport properties varying from insulating to metallic based on oxygen stoichiometry. La and Ti have a relatively high affinity for oxygen. This suggests that the driving force for native oxide formation at the interface between LaTiC >3 and either Ni or Cu should be reasonably low. With an extreme sensitivity to oxygen content, LaTi 03 +x is not particularly attractive as a conductive buffer layer. One possible approach to maintaining metallic conductivity in an oxidized state is through cation doping. This could maintain carrier density as well as reduce oxygen diffusivity. The most likely dopant candidate is Sr. Doping with a divalent element increases the Ti +3 / Ti 4+ ratio and can make the compound less sensitive to ambient oxygen pressure during YBCO deposition. One objective of this work is to investigate the epitaxial growth of intermetallic CuaMg phase as an oxygen diffusion barrier for high temperature superconducting coated conductors. The other one is to grow epitaxial (La,Sr)Ti 03 film as a conductive buffer layer for Cu based RABiTS applications. Chapter 2 reviews the fundamental background of the high temperature superconductors and the rolling-assisted biaxially textured substrate (RABiTS) process including textured metal substrates and buffer layers. In this chapter, preventing method of Cu oxidation, thin film techniques for oxides growth and overview of various conductive oxides for HTS coated conductors will be mentioned. Chapter 3 gives the results of epitaxial C^Mg growth as an oxygen diffusion barrier for Cu RABiTS. Chapter 4 describes the fundamental characteristics of the (La,Sr)TiC >3 film as a conductive buffer layer for coated conductors. The oxygen sensitivity of (La,Sr)Ti 03 film will be investigated with the comparison of LaTiCL film. The electrical transport property such as resistivity as a function of temperature will be covered with various oxygen

PAGE 18

3 pressure conditions. Chapter 5 is for the application of (La,Sr)Ti 03 film as a conductive buffer layer in Ni-W based RABiTS process. In this chapter, epitaxial TiN layer deposited by PLD method on Ni-W tape will be mentioned as a seed layer for (La,Sr)Ti 03 film. Chapter 6 is for the application of (La,Sr)Ti 03 film as a conductive buffer layer in Cu based RABiTS architecture. Epitaxial Ir layer grown by PLD method on Cu tape will be described as an oxygen diffusion barrier. From the result of chapter 6, the feasibility of (La,Sr)TiC >3 film as a conductive buffer layer for Cu based coated conductor will be confirmed. Finally, chapter 7 is the summary for my dissertation.

PAGE 19

CHAPTER 2 LITERATURE REVIEW 2.1 High Temperature Superconductors Superconductivity is an electronic phase transition in which a metal displays a dc resistivity of identically zero and behaves as a perfect diamagnetic material in excluding magnetic field. Most metals do not exhibit a superconducting phase. The low temperature superconductors (LTS) show zero electrical resistance usually well below 20 K. Single elements such as Hg (superconducting transition temperature T c : 4.1 K), Pb (T c : 7.2 K), Nb (7^ : 9.2 K) and various compounds show superconducting behavior at low temperatures. M^Sn (T c : 18.1 K) known as A15 superconductors [1], Li[+ x Ti2x 04 (T c : 13.7 K) [2] and BaPbi_ x Bi x 03 ( T c : 1 1.7 K) [3] are the examples of compound LTS. The BCS theory (proposed by John Bardeen, Leon Cooper and Robert Shrieffer in 1957) gives an explanation of superconductivity behavior [4], According to this model, pairs of electrons (Cooper pair) interact with crystal lattice vibrations in such as way so that an attractive potential results from this interaction. The electron pairs have a slightly lower energy than unpaired electrons and produce an energy gap which decreases from about 3.5kB 7c at T=0 K (where ke is Boltzmann constant = 8.616x10 0 eV/ K) to zero at 7c. This reduced energy gap as a function of temperature was experimentally supported by Townsend et al. [5]. Figure 2-1 is schematic drawing of the energy gap of superconductor material and the energy gap dependence on temperature with theoretical value vs experimental results for some elemental metals [6]. There are two important characteristic lengths associated with superconductivity. One is the London penetration 4

PAGE 20

5 depth, which describes the magnetic field decay inside the superconductor [7]. The penetration depth, A-l, is defined by B z = B z (0) exp(-x/A.| ), where B z is the magnetic field inside the superconducting material, B z (0) is the magnetic field at the interface and x is the distance from the interface. The exclusion of magnetic field by superconductors, known as the Meissner effect, can be explained by this London theory. The other characteristic length is the coherence length, which is related to the transition layer thickness from superconducting state to a normal state. The coherence length is the decay distance of the superconducting wavefunction [8], In addition to T c , there are two more critical parameters that characterize superconducting behavior. The critical magnetic field, H c , is the magnetic field strength above which the superconductivity disappears. It depends on the temperature as described by H c ( T) = H c ( 0) [1 (T / 7Â’ c ) 2 ]. Superconductivity also vanishes if too much current is flowing through the material. The limiting current is called the critical current / c . Superconductivity remains only when the temperature, magnetic field and current are below these three limiting factors ( T c , H c , I c ). Figure 2-2 shows the phase space within which the superconductivity remains (T-H-I diagram). A sharp transition from the superconducting state to a normal state with applied magnetic field is observed in type I superconductors. Practical limitation exists in Type I superconductor because of the low H c value. Type II superconductors has somewhat different transition behavior. The transition from the superconducting state to a normal state is gradual, and there exists an upper critical magnetic field H c 2 which is higher than H c \. The type I and II superconductors can be distinguished by using the Ginzburg-Landau parameter, k, which is the ratio of penetration depth Ai, to coherence length ^(k = / Q. A Type I

PAGE 21

6 superconductor shows the k value far below 1 (£, » A-l). Pure, elemental superconductors are in this category. When A.i, is larger than c, ( k » 1), such superconductors are in the type II category. The magnetic field begins to penetrate through the type II superconductors when the field strength reaches a lower critical magnetic field (H c i). This phenomenon is described as a vortex state. Type II superconductors have cylindrically symmetric domains called vortices which are in the normal state and surrounded by the superconducting matrix. An external magnetic field can penetrate through these vortices which the material remains in the superconducting state.A super conducting current can be maintained in the superconducting matrix if the vortices are pinned to their positions (fluxoid pinning). As the temperature or the magnetic field is increased, the vortex density increases, with vortices getting close to each other. Finally the superconductivity disappears at the upper critical magnetic field (H c 2 ). Figure 2-3 shows the vortex state or mixed state which exists in the type II superconductors. Superconducting alloys and compounds show type II superconductivity with relatively high transition temperatures, flowing large currents and often operating in large magnetic fields [9]. Type II superconductors are sometimes referred to as hard superconductors because of these properties mentioned. A high temperature superconductor (HTS) typically describes a material in which the superconducting transition temperature is greater than 30 K. In 1986, Bednorz and Muller discovered Ba-La-Cu-0 system which showed superconductivity in the 30 K range [10]. Various oxide compounds have been explored since this first HTS discovery. Y-Ba-Cu-0 compounds showing superconductivity transition between 80 and 93 K were

PAGE 22

7 reported by Wu et al. in 1987 [11]. Maeda et al. discovered the Bi-Sr-Ca-Cu -0 system with the T c of about 105 K in 1988 [12]. TheTl-Ba-Ca-Cu -0 superconductor, which showed 120 K transition temperature, was discovered by Sheng et al. in 1988 [13]. The highest T c (above 120 K) was reported in the Hg-Ba-Ca-Cu -0 system by Putilin et al. in 1993 [14]. The generic structure of all HTS compounds consists of layered CuC>2 planes and charge reservoir blocks in a unit lattice cell [ 15 ]. The Cu(>2 planes are separated by divalent or trivalent atoms. The superconductivity, as well as charge transport, are mostly confined to the Cu 02 planes. Figure 2-4 is the schematic drawing of I^CuO,*. Carrier doping by substitution of alkaline earth atoms with trivalent rare earth atoms such as (La,Sr)2Cu04 makes it superconducting at 40 K. YBa2Cu3C>7 has been extensively explored for superconducting devices and wires. YBa2Cu307_g is a hole-doped superconductor possessing a Cu 3+ / Cu 2+ valence state mixture. Its crystal structure is orthorhombic with a = 3.82A, b = 3.88A, and c = 1 1 .68A [16]. Figure 2-5 is the schematic drawing of the YBa2Cu307.g unit cell which can be thought of as an oxygen deficient perovskite structure. One unit cell of YBa2Cu3C>7-g contains one Y atom, two Ba atoms, three Cu atoms and seven O atoms. The YBa2Cu307_g unit cell consists of two Cu 0 2 planes separated by a Y atom. CuO chains are between the Ba -0 layers. Growth temperature and oxygen stoichiometry affects the oxygen site occupancy and transition temperature of the YBa2Cu30 7 _g compound. For films, the lattice parameters can depend on growth temperature. The T c can be reduced below 40 K with 5 = 0.7 [17, 18]. In the case of 5 = 1 , YBa2Cu3C>6 compound is not superconductor and shows a tetragonal structure with a = 3.857A and c = 1 1.819A [19]. Figure 2-6 is the schematic drawing of

PAGE 23

8 YBa2Cu306 compound. The structural difference between 8 = 0 (Figure 2.5) and 8 = 1 (Figure 2-6) is in the CuO chains which play a crucial role in superconductivity. The highest J c can be obtained for current flow parallel to the a-b planes, reflecting the anisotropic crystal structure characteristics. The c-axis oriented epitaxial YBa 2 Cu 307 films can be deposited by various techniques. Coevaporation is one of the earliest proposed methods for YBa 2 Cu 3 C >7 film growth [20-22]. Y, Cu, and BaF 2 from separate sources are coevaporated and an ex situ annealing in oxygen ambient is performed. The RF magnetron sputtering method [23] and pulsed laser deposition method [24] for YBa 2 Cu 3 C >7 film growth were also reported. The detailed thin film growth technique will be explained in Chapter 2.3. The maximum superconducting transition temperature of the Bi-Sr-Ca-Cu-0 system was found at 105 K which is greater than that of YBa 2 Cu 3 C> 7 . The Bi-Sr-Cu-0 system without Ca has a maximum T c value of 22 K [25,26]. The crystal structure of BiSr-Ca-Cu-0 system is an incommensurate superstructure based on an orthorhombic subcell with a = 5.414A, b = 5.44A, c = 30.78A [27], Figure 2-7 is the schematic drawings of Bi-Sr-Cu-0 and Bi-Sr-Ca-Cu-0 compounds. If there is no calcium [Figure 27(a)], the compound shows low temperature superconductivity. The important difference between Figure. 2-7 (b) and (c) is the number of Cu-0 sheets. The compounds of double Cu-0 sheets have orthorhombic c-axis of about 30A, and the compounds that have triple Cu-0 sheets show c-axis of about 37 A [28, 29], This structural difference generated from the number of Cu-O sheets leads to the transition temperature alteration. The compound with the double Cu-0 sheets shows a lower T c phase of around 80 K, but the compound with triple Cu-0 sheets shows a higher T c phase as shown in MaedaÂ’s R-T curve [12].

PAGE 24

9 The three basic high temperature superconductors such as La-A-Cu-0 (A : Ba, Sr, Ca), R-Ba-Cu-0 (R : rare earth element), and Bi-Sr-Ca-Cu-0 compound have Cu-0 sheets where Cu is in essentially square-planar coordination with Cu 2+ . Subramanian suggested that the superconducting mechanism of the Bi-Sr-Ca-Cu-0 compound is related to the orthorhombic distortion of pseudo-tetragonal sheets. Lowering the symmetry of the copper-oxygen sheets and the Bi-0 sheets may play a role in this distortion [30], The bulk Bi-Sr-Ca-Cu-0 compounds can be synthesized by the procedure of calcining, grinding, pressing and sintering in air or oxygen ambient. Bismuth oxide (Bi 2 C) 3 ), strontium carbonate (SrCCfi), calcium carbonate (CaCCL) and copper oxide (CuO) mixture can be used as the synthesizing agents. In order to produce Bi-Sr-Ca-Cu-0 thin films for various device applications, laser sputtering [29], laser ablation [31], successive laser ablation with N 2 O gas [28] and laser molecular beam epitaxy [32] methods have been used. Lead (Pb) can be added to this compound to reduce the processing temperature and increase the transition temperature. In the study of the new cuprate superconductors, Kawai et al. observed that the basic structure units in Bi 2 Sr 2 CaniCu n 02n+4 and Tl2Ba2Ca n -iCu n 02 n +4 consist of layers of Ca(Sr)Cu02 and n=l compound [32]. For example, the mixture of CaCu0 2 with B^S^CuC^ (n=l) gives the compound of Bi 2 Sr 2 CaCu20x (n=2, so-called Bi-2212), and the mixture of 2(CaCu0 2 ) with Bi 2 Sr 2 Cu06 (n=l) gives Bi 2 Sr 2 Ca 2 Cu 30 io (n=3, so-called Bi-2223). For this reason, CaCu0 2 is considered the parent structure of high temperature superconductors. 2.2 Superconducting Power Applications using HTS Wire High-current power transmission cables are among the most exciting opportunities in the high temperature superconductor applications. The operating temperature of HTS materials such as Y-Ba-Cu-0 and Bi-Pb-Sr-Ca-Cu-0 compounds at liquid nitrogen of 77

PAGE 25

10 K certainly has the attractive advantages. The cost savings of the liquid nitrogen is at least 50 times compared to liquid helium which is used to cool the LTS materials, such as Nb3Sn and NbTi, to 4.2 K. In addition, the cryogenic cooling and vacuum insulation systems will be simple by using liquid nitrogen. Other application examples of the superconductors include the field magnet for the motor, the rotor coil for a generator, transformer for power grids, a fault-current limiter which protects a power transmission and distribution system from surges, a current lead for reducing the heat loss in cryogenic machine, magnetic bearings which would be used with large flywheels, magnetically levitated trains for transportation, magnetic resonance imaging (MRI) for medical diagnostic instrument, and superconducting quantum interference devices (SQUID) for magnetic field sensors [33]. First-generation (1G) HTS wire using Bi-Pb-Sr-Ca-Cu-O compounds has the disadvantage of high cost because it is manufactured with superconductive filaments in a silver matrix using the powder-in-tube (PIT) procedure. The detail process of PIT will be mentioned in Chapter 2.3. In order to produce cost-effective long length wire, the process using superconducting Y-Ba-Cu-0 material grown on metal tape has been proposed which is known as the second-generation (2G) wire. An important factor that can influence the current density, J c , in superconducting wire applications is the crystallinity of the HTS material. The high-angle grain boundaries generated in polycrystalline HTS reduce the critical current. The grain boundary can act as a weak superconducting interface. These are known as weak links. The grain boundary effect was demonstrated by Chaudhari et al. [34], For this reason, the method of epitaxial YBa 2 Cu 3 C >7 film deposition on biaxially textured metal tapes was

PAGE 26

11 introduced [35]. The YBa 2 Cu 307 film grown on textured metal templates can drastically reduce the misorientation of the individual grains allowing improvement of the links in the current path. The metal tapes produced by the thermomechanical texturing are known as rolling assisted biaxially textured substrate (RABiTS). The RABiTS process includes depositing buffer layers and HTS materials on the biaxially textured flexible metal substrates. 2.2.1 Biaxially Textured Metal Substrates The basic procedure for RABiTS consists of cold rolling a metal bar in long lengths and subsequent annealing in reducing condition. The primary metals of interest are Ni, Cu and the alloys. The { 1 00}<1 00) cube texture of the metal tape has cube plane parallel to the plane of the sheet and a cube edge parallel to the rolling direction. By continuous rolling, a smooth surface of the metal substrate can be obtained with root mean square (rms) roughness of -50 nm [36]. Subsequent annealing of the as-rolled metal tape enables the sharp {100}(100) cube texture. The annealing is performed at 800-1 000°C in vacuum or ambient of Ar/H 2 mixture gas for a few hours. Figure 2-8 shows the conceptual drawing of the RABiTS process including the buffer layer deposition [36]. The degree of texturing in metal substrates can be measured by x-ray diffraction. Figure 2-9 shows the x-ray co-scan of the (002) Ni which indicates the Ni substrate formed by RABiTS process has out-of-plane alignment. This measurement is done by fixing the sample in the 20 position of (002) Ni and scaning the 0 angle. Figure 2-10 shows the x-ray <|>-scan of the (1 1 1) Ni which indicates the substrate has in-plane alignment. This measurement is done by fixing the sample in the 20 and q/ position of (1 1 1) Ni, and scanning the angle is necessary to observe the in-

PAGE 27

12 plane-alignment. The 20 angle of (002) Ni is 51.844° [37]. The vy angle of (1 1 1) Ni is 35° because the Ni has face-centered-cubic (fee) crystal structure. The full-width-halfmaximum (FWHM) of coand <()-scan graph is calculated by fitting Gaussian distribution to the measured data points. For the sample shown, the FWHM values of coand <)>-scan are 5.4555° and 7.7762°, respectively. These are well matched with the typical data of Ni based RABiTS [38]. Figure 2-1 1 is the log-scale (1 1 1) Ni pole figure of the Ni tape. The pole figure is obtained by fixing the 20 angle of (1 1 1) Ni and rotating the sample by 360° at an individual \\i angle from 0° to 90°. The cube fraction of textured substrate can be calculated by summing the intensities at the cube orientation locations and dividing by the total integrated intensity in the log-scale pole figure. The cube fraction for the sample in figure 2.1 1 is 98.2694%. The four dark circles indicate a well-developed, single component cube texture. The influence of the annealing on the cube texture can also be observed by the x-ray diffraction method. Figure 2-12 shows the x-ray diffraction 0-20 scan graphs of textured Cu substrates before and after annealing. The annealing condition is 800°C in vacuum for 2 hours. The 0-20 scan can not tell any differences between the before and after annealing samples because it only finds every 20 angles that satisfies the Bragg’s condition (nA. = 2d-sin0, here X is the wave length of incident x-ray, d is the inter-plane distance and 0 is the scattering angle). The 20 angle of (002) Cu is 50.4309° [37], On the other hand, the pole figure graphs designate clear differences of cube texture between the two treatments. Figure 2-13 is the log-scale pole figure of (1 1 1) Cu formed by RABiTS process before and after annealing. Without annealing, there are many satellite intensities between each 4 peaks which indicates that the texture is not clearly aligned. After annealing at 800°C in vacuum for 2 hours, the satellite peaks disappear and

PAGE 28

13 the cube fraction was 97.2784% which is higher than the sample without annealing of 84.6100%. Pure Ni is ferromagnetic. This contributes to ac losses. Alloying Ni with W and Fe [39], W only [40] or Cr [41] were suggested to overcome this ferromagnetic problem. Addition of alloying element also increases the yield strength of the metal tape which can be acceptable for a large number of applications. Addition of 3 at. % W can enhance the yield strength of pure Ni from 34 MPa to 150 MPa and reduce the Curie temperature from 627 K to around 400 K [42], 2.2.2 Buffer Layers The first generation of high temperature superconducting wire technology was led by powder in tube (PIT) process with Bi-Pb-Sr-Ca-Cu-0 superconductor material. On the other hand, the second generation (2G) wire is being developed by the RABiTS process with Y-Ba-Cu-0 material. The RABiTS process consists of a biaxially textured metal substrate such as Ni, Cu or their alloys, buffer layers and YBa 2 Cu 307 . The growth of HTS directly on the metal substrate has several obstacles. Due to the oxygen ambient ( ~ 10' 1 Torr) at elevated temperature ( ~ 800°C) during the HTS film formation, the metal substrate can be oxidized resulting in metal oxide on the metal surface. Another problem is the cation substitution in the HTS material. For these reasons, buffer layers between the metal substrate and HTS material have an important role. The buffer layers should prevent oxidation of metal substrate and metal diffusion through the HTS material during the high oxygen pressure and high temperature process. In the case of forming metal oxide on metal substrate, epitaxial layer growth can be impeded which affects the epitaxial layer formation of HTS film. As mentioned in the previous section, the realization of an in plane and out-of-plane aligned HTS film is crucial for achieving high

PAGE 29

14 critical current density J c . The metal oxide also affects the mechanical strength of final coated conductors. When selecting proper buffer layers in the RABiTS process, one should consider the oxidation of metal substrate during buffer layer deposition itself. This is because the process condition of buffer layers can be at high temperature and in an oxidizing ambient. The solution to this problem is to find stable buffer layer materials relative to NiO, CuO or CU 2 O in the growth condition. Jackson et al. reported the thermodynamic stability curves of NiO, CuO and CU 2 O compared to several oxides by calculating Gibbs free energies of the reactants and products [43]. According to this report, MgO, Ce 02 , Y 2 O 3 are good possible candidates for buffer layers. Another important role of buffer layers is chemical separation of HTS from the cation contamination. YBa 2 Cu 307 has proved to accommodate various cationic and anionic substitutions [44], Among these, substitution of Cu with metallic elements dramatically affects the nature of the high-7’ c superconductivity. In the YBa 2 (Cui. x M x ) 307 -§ system where M = Ti, Cr, Mn, Fe, Co, Ni, Ga and Zn, 3 ~ 10% of metal contamination can reduce the T c value below the boiling temperature of liquid nitrogen [45~49]. Table 2-1 shows the T c values reported for YBa 2 (Cu 1 -xMxfiCffg systems. The third important aspect of buffer layers is the mechanical stability and proper adhesion to the metal substrate. For this reason, the lattice constant of the buffer layer should provide a reasonable match with both the metal substrate and HTS materials. The thermal expansion coefficient of the buffer layer also needs to be considered. Buffer layer materials has also been studied for the microelectronic applications of superconducting devices such as superconductor-normal metal-superconductor (S-N-S) junctions or superconductor-insulator-superconductor (S-I-S) tunnel junctions. In these

PAGE 30

15 applications, the starting substrates were single crystal materials such as Si, sapphire (AI2O3), GaAs, MgO or SrTi0 3 . It is difficult to grow high temperature superconducting thin films directly on these substrates because of interdiffusion and the lattice mismatch between them. Several buffer layers were explored for the superconducting microelectronic devices such as MgO [50-52], SrTi0 3 [53-55], Y 2 0 3 [56,57], Yttriastabilized zirconia (YSZ) [58-60], Ce0 2 [61], Zr0 2 [62-64], In the RABiTS application for 2G wire technique, YSZ / Ce0 2 multilayer buffer scheme on Ni tape was proposed by Norton et al. in 1996 [65]. In this architecture, the Ce0 2 layer is the epitaxial template, and YSZ has a role of alleviating oxide cracks. All of the multilayer buffer films were grown by pulsed laser deposition (PLD) method using a KrF excimer laser. After YBa 2 Cu 3 07 formed on this buffer scheme, the superconducting transition temperature (T c ) was observed at 88 K and the critical current density (J c ) for these initial RABiTS structures was 7xl0 5 A/cm 2 at 77 K, 0 T. These results were comparable to those obtained for the epitaxial films on single crystal substrates. In order to retard metal oxidation effectively and overcome the slow deposition rate of PLD method, Pd deposition directly on Ni surface and electron beam evaporation deposition method for the multilayer buffer scheme were proposed. Goyal et al. obtained Jc value of 3x 1 0 5 A/cm 2 at 77 K, 0 T by using PLD deposited YBCO / YSZ / Ce0 2 buffer scheme on ebeam evaporated Pd on Ni tape [35], Paranthaman et al. proposed e-beam evaporated YSZ / Ce0 2 buffer layers on Ni tape with crack free Ce0 2 thin film of 3-10 nm thickness [66], Furthermore, He et al. studied YSZ / Ce0 2 buffer layers on Ni tape by using RF or DC magnetron sputtering method [67], Mathis et al. could get superconducting properties of T c around 88 K, J c exceeding 3xl0 6 A/cm 2 at 77 K, 0 T by adapting RF sputtered YSZ

PAGE 31

16 and e-beam evaporated Ce0 2 multilayer scheme [68], Another buffer scheme using transition metal nitride film was reported by Kim et al. in 2002 [69], In this report, TiN was chosen because of low electrical resistivity (20-30 pQ-cm) and good mechanical strength (YoungÂ’s modulus : 600 GPa, micro-hardness : 2000 Kg/mm 2 ). Due to the oxidation of TiN layer during YBCO deposition, Ce0 2 layer also applied between YBCO and TiN films. The TiN layer was formed by DC reactive sputtering with Ar/N 2 mixture gas and the Ce0 2 , YBCO layers were grown by PLD method. A superconducting transition temperature for this architecture was 89 K and the critical current density was 6x10 A/cm at 77 K. Because TiN is also known as Cu diffusion barrier [70], Cantoni et al. studied multi buffer scheme using TiN for Cu based RABiTS process [71]. In this report, LaMnOa / MgO / TiN buffer architecture was proposed. MgO layer was chosen for oxygen diffusion barrier, and LaMnOa layer proved to be a planarizing material for smooth growing of YBCO. Table 2-2 shows the list of buffer layer materials tested with RABiTS applications referred from [42], 2.3 Review of Conducting Oxide Buffers for Cu based YBCO Coated Conductors 2.3.1 Introduction Since the high temperature superconductors (HTS) were discovered in the late 1980s, worldwide efforts have been made to achieve high-efficiency electric wires. The first generation (1G) multifilamentary HTS wires are composites which can be produced with Bi 2 . x Pb x Sr 2 Ca 2 Cu30io (known as BSCCO-2223) and silver or silver alloy. BSCCO based HTS wire has shown a critical current density above 100,000 A/cm 2 at 77 K [72], Critical current density (J c ) is one of the limiting parameters of superconductivity, above which superconductivity disappears. Bi-Sr-Ca-Cu-O system has a high critical

PAGE 32

17 temperature ( T c ) value as high as 1 10 K with no rare earth elements [73]. The fabrication method of superconducting tapes with BSCCO and silver is known as powder-in-tube (PIT) process. The high-purity mixture of Bi 2 0 3 , PbO, SrC0 3 , CaO and CuO powder is filled into a silver tube and continuously drawn into a narrow cylinder of about 1mm diameter. Such filaments are rolled together and deformed into a tape. Then a final annealing procedure is performed to react and obtain the tapes with good superconducting properties [74]. In order to overcome the high production cost $300/kA-m of 1G wires, YBa 2 Cu 3 07 _g (YBCO) based second generation (2G) wires or tapes are widely being developed in both laboratories and industries [75]. 2G wires have the advantages of lower cost and better electrical performance under applied magnetic fields. The 2G wire with YBCO HTS on RABiTS architecture used by American Superconductor Corporation Inc. (AMSC) is made up of 3 buffer layers scheme on textured Ni-W (5 at.%) tape [76], Y 2 0 3 which can be formed by electron beam evaporation technique serves as a seed layer on 75pm thick Ni-W metal substrate. Yttrium-stabilized zirconia (YSZ) is deposited as a barrier layer. Due to the low level of lattice mismatch with YBCO (-0.12%), Ce0 2 plays a role as a cap layer. Both YSZ and Ce0 2 layers are deposited by an rf-sputtering method. Each buffer layer has a thickness of 75nm. The YBCO film, with the thickness of about 1 pm, is deposited on this buffer stack by metal organic decomposition (MOD) method using trifluoroacetate (TFA) based precursors. Silver is deposited on YBCO film with 3pm thickness for capping the superconductor layer. Finally, a 75pm thick Cu film is bonded for mechanical and electrical stability. The Ag cap layer and Cu stabilizer can also serve as a current shunting path in the event of local defects existing in the YBCO

PAGE 33

18 film. This shunting current cannot flow through the buffer layers because they are all insulators. Figure 2-14 is the schematic drawing of this YBCO based 2G wire architecture, and Figure 2-15 is the SEM cross-section of AMSC’s 2G wire. Nickel, which is the starting template of 2G wires, is ferromagnetic (FM) with a Curie temperature of 63 1 K and a saturation magnetization of 0.5 1 * 1 0 6 A/m at 0 K [77], The magnetic metal substrate, such as Ni, can cause significant hysteretic losses during application of alternating current (ac). The ac loss can be decreased by adding W to Ni [78]. The Ni-W (5 at.%) alloy shows the Curie temperature of 339 K [79] and also increases the yield strength of the substrate to 165 MPa compared to pure Ni substrate of 34 MPa [78]. However, the Ni-W (5 at.%) alloy tape is not nonmagnetic at 77 K. In this aspect, Cu substrate is an attractive candidate for the starting template of 2G wires. Cu is a diamagnetic material that shows no ac loss phenomenon. Cu also surpasses Ni alloy in the electrical conductivity characteristics. At 300 K, the resistivity of Cu tape is 1.5x1 O' 6 Q-cm, which is lower than Ni-W (3 at.%) alloy tape of 2.5x1 O' 5 Q-cm. At the temperature of liquid nitrogen 77 K, Cu tape and Ni-W (3 at.%) tape show the resistivity value of 2.0x10 7 Q-cm and 1.7x10 5 Q-cm respectively. Figure 2-16 is the resistivity curves of Cu tape and Ni-W (3 at.%) alloy tape as a function of temperature. Cu tape has { 100}<100> cube texture and obtained from randomly oriented metal bars by cold-rolling, followed by an anneal in vacuum at 800°C for 1 h. In this 2G wire architecture, the YBCO layer is located in the neutral axis between the 77-78 pm thick Ni alloy substrate with 3 buffer layers, and the 77~78pm thick Ag cap layer with Cu stabilizer. The total thickness of the wire is slightly more than 150pm. For determining the critical current density (J c ), we divide the electrical current value by the

PAGE 34

19 cross-sectional area of YBCO which carries the super current in its superconducting state. The overall engineering critical current density (Je) deals with the current value through the whole cross-sectional area of the wire including not only YBCO but also the substrate, buffer layers, cap layer and stabilizer. One way of increasing the Je value is to reduce the total wire thickness. This can be done by reducing the thicknesses of both the silver cap layer and copper stabilizer, which accounts for half of the total thickness [80]. In this case, the remaining problem is that we have to supply a shunting current path through the starting metal substrate. The current 2G wire structure consists of insulating buffer layers, which means that there is no shunting current path if there is no metallic capping layer on the YBCO film. In order to make the current flow through the metal substrate without capping layer, there must be a conductive buffer layers between the HTS layer and the metal substrate. Compared to Ni or Ni based alloy tapes for 2G wire technology, several profitable aspects can be found in Cu substrates. Cantoni et al. have reported that the thermal stability of coated conductor has the relations with the capping layer thickness and the engineering current density [80]. Figure 2-17 and 2-18 are the simulated graphs of capping layer thickness and engineering current density as a function of the current-width values. The expressions for the capping layer thickness (cf) and engineering current density (Je) are: d c = p c [(K 2 /w) (djps + d m /pm)], J E = K/{d s + d m + Pc [(K 2 /w) (d s /p s + d m /p m )]}, where p c , p s , and p m are the resistivities of capping layer (p c : 2.0x1 O' 7 Qcm for Cu or Ag), superconducting film (p s : 5.0x1 O' 5 Q cm), and metal substrate (p m : 2.0x1 O' 7 Q cm

PAGE 35

20 for Cu and 4.5x 10 7 Q-cm for Ni). K is the current flow per tape width and w is the critical heat flux of the liquid nitrogen at 77 K (w : 10-20 W/cm 2 ) [81]. The thicknesses of superconducting film and metal substrate are designated by d s (d s = K/J s , J s : current density flown through the superconductor) and d m (d m : 50pm). According to these graphs, no capping layer is needed up to K~500A/cm in the case of Cu tape and the maximum Jg can be obtained up to -90 kA/cm 2 for Cu tape which is higher than the value for Ni tape (~60 kA/cm 2 ). The advantages of Cu tape (nonmagnetic, lower material cost (-20% of Ni), easy formation of a sharp cube texture [82] and higher electrical and thermal conductivity (for Cu, k : 398 W/m-K and for Ni, k : 90 W/m-K)) as compared to Ni or Ni based alloy tapes can improve the 2G wire properties along with the conductive buffer layers which offers fully conductive wire architecture. In this review, the possible problems of Cu based RABiTS process and overview of conductive buffer oxides will be addressed. 2.3.2 Preventing Methods of Cu Oxidation In recent research, Aytug et al. demonstrated SrRu0 3 (SRO) / LaNi0 3 (LNO) as a conductive buffer structure for Ni based coated conductors [83], Both SRO and LNO are perovskite type conductive oxides that show metallic behavior. The room temperature resistivity of SRO film is -l^xlO -4 Q-cm and ~3.0xl0‘ 5 Q-cm at 4 K [84]. LNO has a resistivity of 1.8xl0‘ 3 Q-cm at 290 K and 5x\0 A Q-cm at 4.2 K [85], Pulsed laser deposition (PLD) method of YBCO on these conductive buffers is done at high temperatures (~780°C) and high oxygen partial pressures (~2x 10' 1 Torr). These growth condition offer an oxidation environment to the metal substrate. From the work of Aytug et al., Ni substrate oxidation could be observed from XRD and cross-sectional SEM

PAGE 36

21 analysis. Though the Cu based RABiTS process with the conductive buffer architecture has advantages, the oxidation problem could be worse. Figure 2-19 shows the thermodynamic stability curve which indicates that even ~1 *10' 6 Torr of oxygen in the system at YBCO growing temperature can cause copper oxidation. Aytug et al. reported that Lao jSrojMnOs (LSMO) could be used as a conductive buffer layer on Cu based RABiTS applications [86], In this work, they chose a Ni layer on metal substrate to prevent Cu oxidation. However, there are still remaining problems such as ferromagnetism of the Ni layer and NiO formation at the interface between substrate and buffer layer. In metallurgical applications, enhancement of Cu oxidation resistance has been investigated by several approaches, including alloying or implanting elements such as Mg, Cr and Al that are known to be oxygen getters [87-92], Among these, the Mg-Cu and Al-Cu systems are perhaps the most attractive due to the limited solubility of Cr in Cu [93], The use of Mg-doped Cu films as an oxygen diffusion barrier is being investigated for Cu metallization in Si integrated circuit technologies [94]. Unfortunately, the resistivity of Cu increases rapidly with dopant concentration, deterring the use of an alloy as the RABiTSÂ’s substrate material if shunting functionality is desired. However, Mg-Cu or Al-Cu alloy thin film on the Cu substrate may prove to be suitable since the effective electrical transport path length would be limited to a thin surface layer. Another method of preventing Cu oxidation is applying a noble metal as an oxygen barrier between metal substrate and conductive oxide buffer layers. One of the candidate material is iridium (Ir) which has face centered cubic (fee) crystal structure and lattice parameter of 3.840A. In the crystallographic aspect, iridium can be well matched with

PAGE 37

22 Cu substrate. Because iridium is a well known platinum group metal with excellent oxidation resistance, oxidation behavior is only considered above 1400°C [95 — 97]. Although the iridium oxides (Ir0 2 ) are formed after the YBCO deposition, it has good metallic property with resistivity below 3 x 1 O' 4 Qcm at room temperature [98], Paranthaman et al. first tried Ir as an oxygen barrier on Ni-W alloy based RABiTS application [99], In their research, LaojSro jMnC^ conductive buffer layer was used to form fully conductive buffer architecture. On the other hand, careful consideration is also needed, because oxygen diffusivity through Ir is 5xl0 12 cm 2 /s at 800°C [78]. There is also a possibility of Cu diffusion through the Ir layer. Cu and Ir are known to be soluble in extremely small amounts [100], However, 50pm thick Cu substrate can continuously supply copper element to the thin Ir layer during buffer oxides and YBCO processes. These mean that Cu oxidation can be observed either on top of the YBCO film or on the Cu substrate, and proper thickness of Ir layer should be deposited. Another candidate material for preventing Cu oxidation is Pd (palladium). Pd could be a good layer, but there is also an issue with miscibility in the Cu-Pd system. 2.3.3 Thin Film Techniques for Oxides Growth In order to fabricate HTS coated conductors, not only the buffer oxides but also superconducting oxides can be grown by various deposition techniques. These techniques include sputtering, electron beam evaporation, pulsed laser deposition (PLD), and nonvacuum deposition method known as chemical solution deposition (CSD). Sputter deposition is a physical vapor deposition (PVD) technique that has been used widely for thin film growth, especially in the semiconductor industries. Metal films, such as Al, Ti, and W, are sputter deposited for interconnection or metallization

PAGE 38

23 applications. Usually Ar gas is introduced into the vacuum chamber (from few milli Torr to tens of milli Torr of pressure). The positively charged Ar + ions generated by the potential difference between the cathode and anode electrodes bombard the negatively charged target materials. The ejected target elements from these collisions are deposited on the surface of the heated substrate materials. Radio frequency (RF) sputtering is the common way of depositing ceramic materials. Due to the phenomenon of impedance drop of dielectric materials in the high-frequency plasma, the current can flow through the dielectrics. A negatively self-biased target that results from a mobility difference between oscillating electrons and ions enables the insulating ceramic materials to be sputtered [101]. According to the Lorentz force, the path of each charged particle flowing one direction is bent by applying magnetic field. Electrons affected by magnetic field in the plasma move with spiral motion which is called as helix movement. This increases the ion density due to enhanced possibility of collision with gas elements. In other words, deposition pressure can be lowered maintaining the efficiency of sputtering yield. The deposition rate also can be improved with the same applied voltage. These are the advantages of magnetron sputtering. Figure 2-20 shows the schematic drawings of a planar type magnetron sputtering system. In the HTS coated conductor applications various buffer oxides such as YSZ [102], Ce0 2 [103], LaMn0 3 [104], SrRu0 3 , LaNi0 3 [83], Lao 7 Sro. 3 Mn 0 3 [99] are epitaxially grown by the DC or RF sputtering methods. High deposition rate of sputtering system makes it useful for manufacturing long length wires. Electron beam (e-beam) evaporation is one approach to evaporation deposition. In thermal evaporation, materials that needs to be deposited are placed in a crucible and

PAGE 39

24 heated by a resistive heater. The vapor atoms are transferred to the substrates in the vacuum chamber at pressures typically less than l.OxlO " 3 Torr [105], Due to the contamination problem of the crucible material at high temperatures, there is a limitation of using resistively heated evaporation sources. In the case of electron beam evaporation, contamination from the crucible and heating element can be drastically reduced by using shielded heating filaments [101]. The electrons thermionically emitted from this shielded heating filaments are deflected by a transverse magnetic field and reach the surface of the charged material. The evaporation starts by this electron beam heating of the source material. Figure 2-21 is a conceptual drawing of an electron beam evaporation system. Evaporation of MgO [106, 107], Y 2 O 3 [108], CeC >2 [109] are performed by the electron beam evaporation method for HTS buffer layers. Pulsed laser deposition (PLD) is one of the commonly used methods for growing oxides. A KrF excimer is the most popular laser source [1 10]. A KrF excimer laser produces ultra violet (UV) light of 248nm wave length. The laser energy is absorbed by the target material and ablates the target atoms. The ejected atoms generate a plume and travel to the heated substrate material. The distance from the target to the substrate must be considered with respect to the background pressure and ambient gas pressure. Multilayered buffer oxides with various thickness range can be grown by shifting each target material without vacuum break and by changing the repetition rate. Figure 2-22 shows the schematic drawing of a simple PLD system. Due to its simplicity and ablating performance, most of the buffer oxides for FITS application can be deposited by PLD method. Ce0 2 [111], YSZ [1 12], MgO [113], LaMn0 3 [1 13], TiN [113] buffer layers deposited by PLD for HTS application have been reported. The limited laser plume size

PAGE 40

25 and the pulsed deposition mechanism can be a drawback of large scale fabrication of HTS coated conductors. However, high-rate pulsed laser deposition (HR-PLD) method has been proposed by Usoskin et al. [1 14]. Chemical solution deposition (CSD) under non-vacuum condition has great advantages because of low cost and high deposition rate. The general procedures of CSD are synthesis of precursor solution, deposition by spin coating, formation of amorphous film by pyrolysis of organics and crystallization of the coating by high temperature heat treatment [115]. Sol-gel process using 2-methoxyethanol (CH3OCH2CH2OH) as a reagent of metal-oxygen-metal bond formation, and metal organic decomposition (MOD) process using carboxylate compounds as a metal cation source are the common CSD approaches. In the semiconductor industries, sol-gel process is widely used for deposition of TEOS (tetraethyl orthosilicate) oxide as an inter-metallic dielectrics (IMD). BaZr 03 [116], La 2 Zr 207 [117], and Ce02 [118] buffer layers have been investigated by CSD method. Finally, ion beam assisted deposition (IB AD) can be used in the case of polycrystalline metal substrates in order to form the textured template for the buffer oxides. IBAD YSZ and IB AD MgO are the most common templates for producing 2G tapes [119]. 2.3.4 Overview of Conductive Oxides for HTS Coated Conductors An important aspect when considering epitaxial buffer oxides for HTS coated conductors is lattice match with the c-axis aligned YBCO layer. Many of the conducting oxides are in the category of perovskite crystal structure possessing a pseudo-cubic lattice parameter which matches well with the a, b axes of YBCO (a=0.382nm, b=0.389nm). The Perovskite structure consists of two metallic cations and three oxygen atoms (ABO3). The large A atoms occupy each comer of the lattice and oxygen atoms sit on

PAGE 41

26 each face center with small B atoms in the lattice center, completing mostly orthorhombic crystal structure. From another frame of view, the small B atoms are surrounded by 6 oxygen atoms forming BO3 (charge sharing notification) octahedron. This octahedron connects to other octahedra placing large A atom in the middle of their connection. This forms a pseudo-cubic structure. The A and B atoms are selected to make the charge balance. Figure 2-23 is the drawing of a typical perovskite lattice structure. The A site atoms can be cerium (Ce), calcium (Ca), sodium (Na), strontium (Sr), lead (Pb) and rare earth metals. The B site atoms can be titanium (Ti), niobium (Nb), iron (Fe), nickel (Ni), ruthenium (Ru), manganese (Mn), cobalt (Co), chromium (Cr) or copper (Cu) which are normally in the transition metal category. The metallic, insulating and metal-insulator transition properties of a large number of perovskite oxide compounds have been explained by the framework model which includes correlation effects [120, 121]. The band widths W of occupied oxygen 2p states and metalÂ’s d orbitals, the energy difference between oxygen 2p and the lowest unoccupied metal orbital which is designated as A, and the energy difference between the lowest unoccupied metal orbital and the highest occupied metal orbital which is designated as U' are the three terms that can be used in this model. Figure 2-24 is the schematic diagram of the Zaanen, Sawatzky and Allen (ZSA) framework. If the band width W is larger than A or U' as shown in the left and right end side of figure 2-24, the oxides are conductive. The left end side of figure 2-24 designates the overlap of the occupied 2p oxygen valence band with unoccupied metal conduction band, and the right end side designates the overlap of the occupied and unoccupied metal orbitals. In each case partially filled metal orbitals contribute to charge transfer. If UÂ’ is greater than A, the

PAGE 42

27 oxide is called a charge-transfer insulator. If A is greater than U\ the oxide is a MottHubbard insulator. Lanthanum nickelate (LaNiCL) is a conductive perovskite oxide with pseudo-cubic lattice parameter of 0.383 nm. LaNiC >3 is metallic due to the charge transfer gap A closing, which means that the occupied oxygen 2p orbitals and unoccupied 3d Ni orbitals are overlapped. A metal to insulator transition can be occurred in the oxygen deficient phase (LaNi 03 x ) [122], and controlling of oxygen stoichiometry is important. Sputter deposited LaNiC >3 buffered structure (YBCO / LaNiC >3 / Ni tape) and multilayer buffer scheme (YBCO / SrRu 03 / LaNi 03 / Ni tape) are reported as conductive buffer oxides for Ni-based RABiTS [123, 83]. The SrRu 03 layer is used to prevent Ni diffusion through YBCO which can affect degradation of the 7c value. The perovskite SrRu 03 has lattice constants of a = 0.555 nm, b = 0.556 nm, c = 0.786 nm, and a, b lattice constants are close to V2 ayBco or V2 b Y Bco with pseudo-cubic lattice parameter of 0.393 nm. The conductive property of SrRu 03 comes from the overlap between oxygen anion p orbitals and cation t 2g [124], Metallic SrRu0 3 is also ferromagnetic with a Curie temperature of — 160°K [125]. There have been several reports of SrRuC >3 buffer layers as diffusion barriers on Pt [126] and Ni [127-129] for HTS coated conductor applications. (Lai. x ,A x )Mn 03 is the interesting compound system which has both metallic and ferromagnetic properties. A represents a divalent alkali elements such as Sr, Ca or Ba. In the extreme composition of LaMnC> 3 , La and Mn must have 3+ charges in accordance with the 6charge of three oxygen atoms. LaMnC >3 is an insulating antiferromagnet where electrons are localized on the atomic orbitals [130]. When the La atoms are

PAGE 43

28 replaced by divalent elements in the concentration range of 0.23 5 or La 2 Ti 2 0 7 ), it becomes a ferroelectric band insulator. LaTiC >3 has an orthorhombic perovskite structure with a = 5.604A, b = 5.595 A and c = 7.916 A [136]. In contrast with LaTiC> 3 , La 2 Ti 2 0 7 shows monoclinic layered structure. According to the phase diagram study, LaTi 03 +x changes its electric and magnetic properties with oxygen content [137, 138], LaTi 03 +x is an attractive material for coated conductor application because it has good metallic property at all temperature ranges in the oxygen stoichiometric of 0.1 < x < 0.25 [137], and its pseudo-cubic lattice parameter is well matched with YBCO. However, the composition also can be changed during YBCO

PAGE 44

29 growth in the high oxygen ambient (~1 xlO ' 1 Torr) at high temperature (~780°C). The previous research on the relation of resistivity of LaTi 03 +x with oxygen pressure during deposition by PLD indicated that oxygen played a crucial role in conducting property [139], Cation doping in the compound can overcome this oxygen sensitivity of LaTi 03 +x . Doping with divalent elements increases the Ti ' 3 / Ti 4+ ratio and can make the compound less sensitive to ambient oxygen pressure during YBCO deposition. Electrical conductivity is a function of doping constant x in the Lai. x Sr x Ti 03 compound, and the resistivity continuously decreases with higher cation doping [140]. Previous studies on oxygen dependency showed that the room temperature resistivity of Lao sSro .sTiC^ remained below 1.0x10 3 Q-cm in 10 4 ~ 10 ’ 2 Torr of oxygen pressure range [141]. Recently, conductive buffer architecture with PLD deposited (La,Sr)TiC >3 film on Cu based RABiTS application (YBCO / LSTO / Ir / Cu tape) was reported [142], Lai_ x Sr x Co 03 is the compound that has mixed valency due to the cation doping. The parent compound LaCo 03 is an insulator which has a charge transfer gap A (see figure 2.24) formed between an occupied oxygen 2p band and an empty cobalt 3d e g band [132]. By cation doping to LaCo 03 , the ratio of Co 3 + /Co 4+ increases introducing metallic behavior. The transport and magnetic properties depend on the doping concentration. The room temperature resistivity of Lao. 5 Sro. 5 CoC >3 is known to be ~9xl0 ' 5 Q-cm. The pseudocubic lattice parameter of Lao.sSro 5 C 0 O 3 is 3.835A. The oxygen stoichiometry is an important parameter for resistivity control because oxygen deficiency can change the Co 3 + /Co 4+ ratio and the structural disorder in the Co-O-Co conduction channel [143]. LaCuC >3 is a compound which has Cu 3+ ions. In a recent study, hybridization of unoccupied oxygen p states with Cu 3d and La 5d states was suggested as an electronic

PAGE 45

30 structure of LaCuC >3 [144]. The stoichiometric form of La 2 Cu 04 is an insulator. However, the conduction behavior of LaCu 03 _g alters from an insulator to metallic conductor with 8 ranging from 0 to 0.5. Although the nonstoichiometric LaCuOa-g has high conductivity with available oxygen vacancies, it is not a desirable material due to difficulties of synthesis under high oxygen partial pressure. Yu et al. reported that Sr doping in the LaCu 03 _g compound stabilized the perovskite structure. Enhancement of conductivity was observed in Lai. x Sr x Cu 02 -g [145]. The overall electrical and structural properties of perovskite oxide compounds mentioned in this section is briefly shown in table 2-3. The major factor of selecting conductive buffer oxides for HTS coated conductor application should include not only the structural, chemical compatibility with metal substrate and YBCO film but also transport properties that need to be sustained even after the severe oxidizing process during YBCO formation.

PAGE 46

31 Table 2-1. The critical temperature values in absolute temperature unit reported for YBa 2 (Cui.xM x ) 3 07 ^ systems. X Ti Cr Mn Fe Co Ni Zn Ga Ref. 0.1 75 84.5 78.9 38 21.2 66.3 <3 45 0.033 61 55 80 70 51 46 0.1 86 92 50 50 73 40 47 0.033 76 73 55 55 48 0.033 88 88 80 58 49

PAGE 47

32 Table 2-2. The list of buffer layer materials tested with RABiTS applications. Buffer Material Cubic Lattice Parameter (A) % Lattice Mismatch versus YBCO % Lattice Mismatch versus Ni Oxygen Diffusivity 'y (cm /s) at 800°C Electrical Resistivity (p.Qcm) MgO 4.210 9.67 17.74 8x1 O' 22 SrTi0 3 3.905 2.16 10.26 6xl0" n Ce0 2 5.411 0.12 8.22 6x1 O’ 9 Y 2 0 3 10.604 -1.89 6.22 6xlO' 10 YSZ 5.139 -5.03 3.07 2x1 O' 8 TiN 4.242 10.43 18.49 20-30 LaMn0 3 3.880 1.60 9.70 8xl0’ 15 Ni 3.524 5-6 Cu 3.615 1.5-2

PAGE 48

33 Table 2-3. The overall electrical and structural properties of perovskite oxide compounds mentioned in this section. o 2 c 3 CO c-> 9* = < -O c c U a: c 00 8 .S g S & § g CL c ^ o. s o z 3 li E o — ^ O O ' o O O' £ S 2 ^ . E E o o d d o D. O < o < cd VC “ 2 "" ^ X O P u ^ CO pa •: •• ?“• *"} O ^ O tG G cs v> i i C 00 8 .£ « ,u §) § g I CO >• tZ S’ E o E E O O G G — (N J f E o G 5 O a — °< so m vO °^f m °< t-m up Os «/3 rn a -if € 9 f -g 2 3 ? 13 E -J 2t S o P T3 Q 3 5 c 3 ftS 00 i O H c/5 Q Q Q c 9 U

PAGE 49

34 0 " N(E) a i\ 1 \ i i i i ! Eg ] — i i JJ (a) Figure 2-1. (a) Schematic drawing of energy gap in superconducting material, (b) theenergy gap dependence on temperature with theoretical value versus experimental results for some elemental metals.

PAGE 50

35 Figure 2-2. Schematic drawing of the critical space within which the superconductivity remains (T-H-I diagram). Figure 2-3. Phase diagram of type II superconductor with temperature and magnetic field dependency. H +

PAGE 51

36 Cu 0 2 La La CL1O2 La La C11O2 charge reservoir C11O2 charge reservoir CL1O2 Figure 2-4. Schematic drawing of La 2 CuC> 4 .

PAGE 52

37 Yttrium Barium Copper Oxygen Figure 2-5. Schematic drawing of YBa 2 Cu 307-8 unit cell combined with oxygen deficient perovskite structure.

PAGE 53

38 ±= 5 ^ ±=k' ^=k' Yttrium Barium Copper Oxygen Figure 2-6. Schematic drawing of YBa2Cu306 compound which has different CuO chains with YBa 2 Cu 307 .

PAGE 54

39 BiO GF O Bi 0 Sr © Ca O Cu o O m (a) (b) (c) Figure 2-7. Schematic drawing of (a) Bi-Sr-Cu-O, (b) Bi-Sr-Ca-Cu-0 (double Cu-0 sheets), and (c) Bi-Sr-Ca-Cu-0 (triple Cu-0 sheets).

PAGE 55

40 rolling o o o o o o ANNEALING BUFFER LAYER DEPOSTION RABiTS Figure 2-8. Conceptual drawing of rolling assisted biaxially textured substrate (RABiTS) process based on Ni substrate. The pattern inside the circle designates the texture alignment of the substrate.

PAGE 56

41 (D c 15 20 25 30 35 e (deg) Figure 2-9. The X-ray diffraction co-scan (rocking curve) of (002) Ni which indicates the substrate’s out-of-plane alignment. The full-width-half-maximum (FWHM) of the curve is 5.4555°.

PAGE 57

42 (D C -150 -100 -50 0 50 100 150 (deg) Figure 2-10. The x-ray diffraction <(>-scan of (1 1 1) Ni which indicates the substrate’s inplane alignment. The average full-width-half-maximum (FWHM) of the 4 peaks is 7.7762°.

PAGE 58

43 Figure 2-11. The log-scale (1 1 1) Ni pole figure of the Ni substrate formed by RABiTS process. The cube fraction of the substrate is 98.2694%.

PAGE 59

44 (a) 30 35 40 45 50 55 20 (deg) (b) Figure 2-12. The X-ray diffraction 0-20 scan graphs of textured Cu tapes formed by RABiTS process (a) before annealing and (b) after annealing at 800°C in vacuum for 2 hours.

PAGE 60

45 (a) Figure 2-13. The log-scale pole figures of (1 1 1) Cu substrates (a) before annealing and (b) after annealing at 800°C in vacuum for two hours.

PAGE 61

46 YBCO HTS layer -> ~ 1 pm Ce0 2 cap layer -»• 0.75 pm YSZ barrier layer -> 0.75 pm Y 2 0 3 seed layer -> 0.75 pm Figure 2-14. Schematic vertical structure drawing of conventional 2G wire based on YBCO HTS. Figure 2-15. SEM cross-section of AMSC’s 2G wire.

PAGE 62

47 Figure 2-16. Resistivity comparison of Cu tape and Ni-W (3 at.%) tape as a function of the temperature.

PAGE 63

48 Figure 2-17. Simulated graph of cap layer thickness as a function of current per tape width.

PAGE 64

49 Figure 2-18. Simulated graph of engineering current density as a function of current per tape width.

PAGE 65

50 Temperature (°C) Figure 2-19. Thermodynamic stability curve of Cu in terms of temperature and oxygen partial pressure.

PAGE 66

51 Heating Block Figure 2-20. Schematic drawing of magnetron sputtering system.

PAGE 67

52 Substrate Figure 2-21. Schematic drawing of electron beam evaporation system. Electron Beam Chamber Wall

PAGE 68

53 Vacuum Chambe r Ablation Target Scanning Mirror^ Pulsed Laser Beam Ablation Plume Substrate Heater / Backgroud Gas Figure 2-22. Schematic drawing of pulsed laser deposition system.

PAGE 69

54 Figure 2-23. Schematic drawing of typical perovskite crystal structure. semichargeMott-Hubbard metal trasfer insulator insulator U'
PAGE 70

CHAPTER 3 GROWING EPITAXIAL CU 2 MG AS AN OXIDATION BARRIER FOR HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 3.1 Introduction High-temperature Superconducting (HTS) biaxially-textured coated conductors hold significant promise for the development of a superconducting wire technology functional at the liquid nitrogen temperatures (64-77 K). To date, this technology utilizes epitaxial YBa 2 Cu 307 (YBCO) coatings deposited on biaxially-textured Ni or Ni-based alloy substrates [35][39][41][151], These tape substrates are fabricated by thermomechanical deformation of the metals. While HTS wires based on the biaxially textured Ni substrates offer the promise of substantial energy savings for applications in the power sector, the actual impact of HTS wire will be determined by several factors, including the reliability of these conductors against catastrophic failure, as well as the potential limitations introduced via the use of ferromagnetic substrates. One approach that addresses both substrate ferromagnetism and supercurrent shunting is to use high conductivity Cu tapes as the base metal substrate. For use as a substrate for HTS coated conductors, the oxidation of the Cu substrate must be addressed. A significant issue with Cu is the oxidation of the metal substrate due to the following buffer oxide and YBCO depositions at the elevated temperature with high pressure of oxygen gas. The copper oxide (Cu 2 0, CuO) growth occurs at the oxide-gas interface and the rate-determining step of oxidation is the diffusion of cation vacancies [152]. In general, the oxidation of Cu proceeds rapidly and at minimal oxygen partial pressure. At a temperature of 700°C, 55

PAGE 71

56 -8 Cu 2 0 formation is favored for pure Cu exposed to oxygen pressures as low as 10‘ Torr. A typical RABiTS is composed of several buffer layers such as Ce0 2 , Y 2 C >3 and yttriastabilized zirconia (YSZ). At the temperatures of YBCO deposition, these oxides cannot block oxygen diffusion causing the oxidation of Cu substrate. Copper oxide is not an effective passivation layer to further oxidation or scaling. In metallurgical applications, enhancement of Cu oxidation resistance has been investigated by several approaches, including alloying or implanting elements such as Mg, Cr and A1 that are known to be oxygen getters [87]-[92]. Among these, the Mg-Cu and Al-Cu systems are perhaps the most attractive due to the limited solubility of Cr in Cu [93], The use of Mg-doped Cu films as an oxygen diffusion barrier is being investigated for Cu metallization in Si integrated circuit technologies [94], Unfortunately, the resistivity of Cu increases rapidly with dopant concentration, deterring the use of an ally as the substrate material if shunting functionality is desired. However, Mg-Cu or AlCu alloy thin film on the Cu substrate may prove suitable since the effective electrical transport path length would be limited to a thin surface layer. The phase diagram for the Cu-Mg binary alloy system indicates the presence of intermetallic Cu 2 Mg at the range of 15-18 weight % Mg up to 819°C [153]. The crystal structure of Cu 2 Mg is cubic fee with the lattice constant a = 7.064A [154], In addition the room temperature resistivity of Cu 2 Mg is known to be 5-6 pQcm which is compatible with that of Cu (1.67 pQcm) [155], One can therefore consider two variants of (Mg,Cu) buffer layers for Cu substrates, namely Mg-doped fee Cu and the cubic Cu 2 Mg intermetallic. In order to assess the potential applicability of this approach to oxidation resistant buffers for

PAGE 72

57 RABiTS-based conductors, we have investigated the epitaxial growth of (Cu,Mg) films on a (001) Cu surface. 3.2 Experiments For these experiments, epitaxial (001) Cu films on (001) MgO single crystals were used as the substrate materials. The Cu films were grown using sputter deposition at a substrate temperature of 400°C with 1.0x1 O' 2 Torr of Ar gas. The thickness of the epitaxial Cu film was 180nm. Due to the high vapor pressure of Mg (over 2.0x1 0' 3 Torr at 400°C), in situ growth of epitaxial (Cu,Mg) films is not possible. The approach used to achieve epitaxy of a (Cu,Mg) film was to form Cu/Mg multilayer precursor films that would be subsequently annealed to form either Mg-doped fee Cu or intermetallic Cu 2 Mg. Sputter deposition was used to deposit Mg and Cu multilayers at room temperature with ~2 _ 1 .0x 1 0 Torr of Ar gas. The precursor consists of an Mg/Cu multilayer stack with 5 each of 25 nm thick Mg and 25 nm thick Cu layers, which were grown at room temperature by sputter deposition. Figure 3-1 shows the vertical structure, which is used for this experiment. The precursor is then annealed in a flowing F^/Ar mixture at temperatures ranging from 400°C to 700°C. The development of (Cu,Mg) phases as the multilayer was investigated as a function of annealing temperature and duration. 3.3 Results and Discussion At annealing temperature of 400°C, formation of the intermetallic Cu 2 Mg was observed. Figure 3-2 shows a 2-0 X-ray diffraction scan along the surface normal for a multilayer structure annealed at 400°C. The strongest peaks, other than that from the substrate and epitaxial Cu film, correspond to Cu 2 Mg. Both (111) and (001) oriented Cu 2 Mg grains are observed. Interestingly, the (001) oriented component of the

PAGE 73

58 intermetallic phase was found to be epitaxial with respect to the C layer. Figure 3-3 shows the X-ray diffraction rocking curve for the (004) Cu 2 Mg peak, indicating a fullwidth half-maximum (FWHM) of A0 = 2.0°, which is slightly larger than that for the Cu film (A0 = 1 .45°). The in-plane orientation of the c-axis oriented Cu 2 Mg was investigated using four-circle x-ray diffraction. Figure 3-4 shows an in-plane cp-scan through the Cu 2 Mg (222). The four-fold symmetric peaks indicate that the grains are epitaxial with respect to the Cu film, possessing a cube-on-cube orientation. The FWHM of the in-plane peaks is 2.1°. The cubic Cu 2 Mg lattice parameter was measured to be 7.016 ± 0.005 angstroms. At the upper end of the temperature range (600°C, 700°C), only the (001) Cu peaks are observed. This is consistent with the formation of an epitaxial Mg-doped fee Cu alloy. Figure 3-5 is the 2-0 x-ray diffraction scan of the (Cu,Mg) multilayer after annealing at 400, 500, 600, 700°C. In order to confirm the oxidation resistance of the structures possessing (Cu,Mg) alloy films, Ce02 was deposited at elevated temperatures on Ni / (Cu,Mg) / Cu / MgO substrates and compared to CeC >2 films on Ni / Cu / MgO. The CeC >2 films were deposited by PLD at 750°C, and included a thin nucleation layer deposition in vacuum, followed by CeC >2 deposition at 2x1 O' 4 Torr of oxygen. In case of the CeC >2 film on Ni / Cu / MgO, significant surface roughness due to the metal oxidation is observed. This is shown in figure 3-6. In contrast, no surface roughness is observed in the SEM image for the CtO^I Ni / (Cu,Mg) / Cu / MgO structure. 3.4 Conclusions The growth of epitaxial Cu 2 Mg as an oxidation barrier for high temperature superconducting coated conductors was investigated. Epitaxy of (004) Cu 2 Mg intermetallic phase was achieved on (002) Cu film. In-plane (p-scan through the Cu 2 Mg

PAGE 74

59 (222) and the x-ray diffraction rocking curve for the (004) Cu 2 Mg peak indicates that the intermetallic phase is well oriented on (002) Cu surface.

PAGE 75

60 Cu 25nm 25nm Cu 25nm 25nm Cu 25nm 25nm Cu 25nm 25nm Cu 25nm 25nm Cu l$0nm MgO Mg Mg Mg Mg Mg Figure 3-1 . The vertical structure which is used for this experiment.

PAGE 76

intensity (arb) 61 20 (deg) Figure 3-2. The X-ray diffraction 0-20 scan along the surface normal for a multilayer structure annealed at 400°C.

PAGE 77

62 0 (deg) Figure 3-3. The X-ray diffraction rocking curve for the (004) Cu 2 Mg peak, indicating a full-width half-maximum (FWHM) of A0 = 2.0°.

PAGE 78

intensity (arb) 63 4 (|)-scan through the Cu 2 Mg (222) • 0 i i i — i — — i — — i — — i — A4> = 2.1° FWHM (deg) Figure 3-4. The X-ray diffraction <))-scan through the Cu 2 Mg (222).

PAGE 79

64 20 (deg) Figure 3-5. The X-ray diffraction 0-20 scan of the (Cu,Mg) multilayer after annealing at 400, 500, 600, 700°C.

PAGE 80

65 Figure 3-6. The SEM picture of (a) Ce0 2 film on Ni / Cu / MgO (b) Ce0 2 / Ni / (Cu,Mg) / Cu / MgO structure.

PAGE 81

CHAPTER 4 (LA,SR)TI0 3 AS A CONDUCTING BUFFER LAYER FOR HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 4.1 Introduction Among the potential application of high-temperature superconductivity (HTS) are the power and utility sectors that will require high-current wires [8], In the RABiTS (rolling assisted bi-axially textured substrate) approach, HTS coated conductors are made by deposition of buffer layers and superconducting oxide on a crystalline-textured metal tape. This approach is conducive to the development of electrically conductive buffer layers, which would provide a current shunt to the metal tape for improved protection. There are key issues for a conductive buffer layer of HTS coated conductors. First, it must be reasonably well lattice matched to both the metal substrate and the superconducting film, thus enabling epitaxy. Second, the interaction between the buffer layer and the metal substrate must be such as to minimize formation of any native interfacial oxide that would serve as an insulating barrier to shunted current flow. Although buffer layers of the conductive oxides Lao.7Sr 0 . 3 Mn0 3 , SrRu0 3 and LaNi0 3 on textured Ni tape have been reported [83,133], to date most buffer layers investigated for HTS conductors have been insulating, such as Ce0 2 [156-158], Y 2 0 3 -Zr0 2 [68,151], [159,160], Zr0 2 [161], SrTi0 3 [162] and NiO [163], [164]. In the latter situation, protection must be accomplished by the addition of an adjacent normal metal layer that is sufficiently conductive and thick to accommodate the current flow without unmitigated growth of an unstable hot zone. 66

PAGE 82

67 An alternative candidate material system that may satisfy the criteria for conductive buffers mentioned above is (La,Sr)Ti03. LaTi03 +x is an interesting defect perovskite system, with transport properties varying from insulating to metallic based on oxygen stoichiometry [138,140,165]. La and Ti have a relatively high affinity for oxygen. This suggests that the driving force for native oxide formation at the interface between LaTiCL and either Ni or Cu should be reasonably low. In previous work, we have demonstrated epitaxial growth of LaTiC>3 on (001) Ni using pulsed laser deposition [166], The results indicate that epitaxy can be achieved, although the stability of the LaTiC^/Ni structure is limited to reducing conditions due to the phase transitions that occur with increased oxygen content. With an extreme sensitivity to oxygen content, LaTiC>3 +x is not particularly attractive as a conductive buffer layer. One possible approach to maintaining metallic conductivity in an oxidized state is through cation doping. This could maintain carrier density as well as reduce oxygen diffusivity. The most likely dopant candidate is Sr. Doping with divalent element increases the Ti 3 / Ti 4+ ratio and can make the compound less sensitive to ambient oxygen pressure during YBCO deposition. Electrical conductivity is a function of doping constant x in the Lai_ x Sr x Ti03 compound, and the resistivity continuously decreases with higher cation doping [140], Previous study on oxygen dependency showed the room temperature resistivity of Lao sSro sTiCL was remained below 1.0x10 3 Q cm in 1 0" 4 ~ 10Â’ 2 Torr of oxygen pressure range [141]. In this chapter, the fundamental properties of (La,Sr)Ti0 3 films on single crystal SrTi03 substrates have been investigated and compared with the LaTiC>3 films. The HTS films such as YBa2Cu3(>7 grown on (La,Sr)Ti03 buffer layers on single crystal SrTiC>3

PAGE 83

68 substrates have been also investigated in order to evaluate the compatibility of thin film layers. 4.2 Experiments The growth of (La,Sr)Ti 03 and LaTiC >3 films on (001) oriented single crystal SrTiC >3 substrate, 0.2-0.5p.m thickness, was performed by pulsed laser deposition (PLD) in vacuum at 750°C, at an energy density of 2 J/cm 2 and repetition rate of 10Hz. A KrF (248nm) excimer laser was used as the ablation source. The oxygen sensitivities of (La,Sr)TiC >3 and LaTiC >3 films have been investigated by changing the oxygen partial pressure from 3.0xl0? Torr to 4.0xl04 Torr. The high temperature superconducting (HTS) YBa 2 Cu 3 C >7 (YBCO) layer was also grown by PLD under conditions of 1.0x1 O' 1 Torr of oxygen at 780°C, energy density of 2 J/cm 2 , 10Hz rate, with a thickness of 0.2pm. After completing YBCO deposition, the PLD chamber was cooled down to 500°C at a rate of 28°C/min and the oxygen pressure was increased to 400 Torr. After 20 minutes, the chamber was cooled down to room temperature rapidly. The x-ray diffraction of 9-20 scan, co and (j)-scan were used in order to observe the thin film crystallinity, in-plane and out-of-plane alignments of each film layers. The in situ reflection high-energy electron diffraction (RHEED) was used to monitor epitaxial film growth. The film thickness was measured by step profilometer which converts the physical step distance into the electrical signals. A standard four-point probe technique was used to evaluate the electrical properties, including temperature-dependent resistivity (R-T) of not only the buffer layers but also the superconducting film, the superconducting transition temperature ( T c ) and the critical current density ( J c ). The R-T measurements were done in the cryogenic cooling

PAGE 84

69 system which was operated in the liquid helium of 4 K. The critical current density (J c ) as a function of magnetic field was evaluated by magnetization method in liquid helium. The widths of the samples were in the range of 0.28 ~0.53cm and the distance between the voltage tips was in the range of 0.3~0.4cm. 4.3 Results and Discussion Figure 4-1 shows the x-ray diffraction 0-20 scan of (La,Sr)Ti 03 film deposited on (001) oriented SrTiC >3 single crystal substrate. The film was grown by pulsed laser deposition (PLD) method at 750°C in vacuum. The y-axis which indicates the intensity of figure 4-1 is designated by log scale in order to separate the film and substrate peaks clearly. The x-ray 0-20 scan of (La,Sr)Ti 03 film shows (001), (002), (003) peaks at 22.28°, 45.50°, and 70.93° respectively. The cubic lattice constant of the film is 3.98A calculated from the Bragg’s law. Figure 4-2 is the in situ monitored reflection highenergy electron diffraction (RHEED) pattern of (La,Sr)Ti0 3 film grown at 750°C in vacuum. According to the x-ray 0-20 scan and RHEED pattern, the (La,Sr)Ti 03 film is epitaxially deposited on SrTi 03 single crystal substrate by PLD method at 750°C in vacuum. In order to examine the oxygen sensitivity, the (La,Sr)Ti 03 films were grown in oxygen ambient with partial pressure of 3.0xl0' 5 Torr and 4.0x1 0 -4 Torr. Figure 4-3 shows the x-ray diffraction 0-20 scan of (La,Sr)Ti 03 films deposited on SrTiC >3 single crystal substrates by PLD method with different ambient conditions. Only the (002) peaks of each sample are designated in this figure to make clear comparison. The dashed lines indicate the standard SrTi 03 (002) and LaTi 03 (002) peak positions at 46.48°, 46.03° respectively. The films grown under the oxygen partial pressure show the (002) peaks at almost the same position with LaTi0 3 (002) at 46.03°. This indicates that the

PAGE 85

70 oxygen deficiency in the film is compensated by ambient oxygen pressure leading to the stoicheometric LaTiC >3 film. Figure 4-4 shows the x-ray diffraction 0-20 scan of LaTi 03 film deposited on (001) oriented SrTiC >3 single crystal substrate. The film was grown by pulsed laser deposition (PLD) method at 750°C in vacuum. The x-ray 0-20 scan of LaTi0 3 film shows (001), (002), (003) peaks at 22.26°, 45.47°, and 70.89° respectively. The cubic lattice constant of the film is calculated by 3. 98 A. In order to compare the oxygen sensitivity of LaTiC >3 with (La,Sr)TiC >3 film, same oxygen partial pressure conditions such as 3.0x1 O' 5 Torr, 4.0x1 0" 4 Torr were applied during LaTiC >3 deposition. In the figure 4-5 (a) and figure 4-6 (a), the x-ray peaks were taken with aligning the sample to the (001) oriented SrTiC >3 substrate. There are no clear peaks which designate the film existence in both cases. In contrast with LaTiC> 3 , La2Ti20? which is a fully oxidized phase shows monoclinic layered structure. The crystallographic angle between (-210) La 2 Ti 207 and (001) SrTiC >3 is known to be 4.52° [4.13], The 0-20 scan graphs in the figure 4-5 (b) and figure 4-6 (b) were taken by tilting the substrate around 4° off axis. According to the figure 4-5 and figure 4-6, LaTiC >3 films which are growing under the oxygen ambient can be turned into the fully oxidized L^T^CF phase. Figure 4-7 shows the x-ray diffraction <|>-scan of (-420) peak for the L^T^CT layer grown under the oxygen pressure of 3.0x1 0’ 5 Torr. The fourfold symmetry of the peak indicates that the film is inplane aligned. In addition to the structure, the transport behavior of epitaxial (La,Sr)TiC >3 and LaTiC >3 films grown on SrTiC >3 is indicated in figure 4-8 as a function of the oxygen pressure during growth. Figure 4-8 (a) designates the resistivities measured at 300 K and figure 4-8 (b) shows the resistivities measured at 77 K. In case of the vacuum growth

PAGE 86

71 condition, both (La,Sr)Ti03 and LaTi03 films show low and nearly identical values, with a resistivity on the order of 10 Q cm at 300 K. As oxygen pressure is moderately increased up to 4.0x1 0" 4 Torr, the resistivity of LaTiCh film increases by more than two orders of magnitude to ~0.2 Q cm at 300 K. This increase in the resistivity with oxygen pressure during growth is related to the structural transition from LaTiC>3 to La2Ti207 phase as indicated in the X-ray diffraction pattern of figure 4-5 and figure 4-6. In contrast, the resistivity of (La,Sr)TiC>3 films is relatively insensitive to oxygen pressure making it attractive as a conductive buffer for coated conductor applications. Figure 4-9 and 4-10 show the resistivity curves of (La,Sr)Ti0 3 and LaTi0 3 films on SrTi0 3 substrates under various oxygen pressures as a function of measuring temperature. All of the R-T curves indicate that the films show metallic behaviors. Based on the fundamental results of (La,Sr)Ti03 films on single crystal SrTiC>3 substrate, high temperature superconducting YBa2Cu3C>7 layer was deposited on (La,Sr)Ti03. In this case, the (La,Sr)TiC>3 layer played a role of buffer layer similar to the real coated conductor application such as RABiTS. Figure 4-11 is the X-ray diffraction 020 scan of YBa2Cu307 film grown on (La,Sr)Ti03 buffer layer on single crystal SrTiCh substrate. The (00/) peaks of YBa2Cu3C>7 film are well defined. Figure 4-12 shows the Xray diffraction co-scan of (002) peak for (La,Sr)TiC>3 buffer layer grown on SrTi03 substrate and (006) peak for YBa2Cu3C>7 HTS film deposited on (La,Sr)TiC>3 buffer layer. The full width half maximum (FWHM) values of each layer are 0.97° and 1.58°, respectively. Figure 4-13 shows the X-ray diffraction 3 buffer layer grown on SrTiC>3 substrate and (012) peak for YBa2Cu3C>7 HTS film deposited on (La,Sr)TiC>3 buffer layer. The average full width half maximum

PAGE 87

72 (FWHM) values of the fourfold symmetric peaks for each layer are 0.71° and 1.48°, respectively. According to the co-scan and <)>-scan, the YBa 2 Cu 3 0 7 film is in-plane and out-of-plane aligned to the (La,Sr)Ti0 3 buffer layer. Figure 4-14 shows the four-probe resistivity measurement for the YBa 2 Cu 3 0 7 / (La,Sr)Ti0 3 / SrTi0 3 structure. The sample has the superconducting transition temperature (T c ) of 91 K indicating cationcontamination-free YBa 2 Cu 3 0 7 . Figure 4-15 is the critical current density (,/ c ) as a function of the magnetic field of YBa 2 Cu 3 0 7 / (La,Sr)Ti0 3 / SrTi0 3 structure. The J c values were evaluated by magnetization method in liquid helium. The J c can be calculated by the equation of J c = (15xAM) / r, here AM is the magnetization difference and r is the radius of the sample. In this graph, the zero-field transport J c value is 2.18xl0 6 A/cm 2 . The interesting aspect of the conductive buffer layer in the coated conductor application is whether its conductivity remains after HTS film deposition or not. Normally, the HTS film such as YBa 2 Cu 3 0 7 is grown at high temperature in high oxygen pressure. The underlying buffer layer should be exposed to this extreme oxidizing condition. The resistivity of (La,Sr)Ti0 3 film after annealing at YBa 2 Cu 3 0 7 deposition condition (at 780°C, P 02 l.OxlO" 1 Torr). Figure 4-16 shows the resistivity curve of (La,Sr)Ti0 3 film grown on SrTi0 3 substrate after annealing at YBa 2 Cu 3 0 7 deposition condition. The resistivity values at 300 K and 77 K were about l.OQcm and 0.09 Q cm. The resistivity values of (La,Sr)Ti0 3 film without annealing were 3.5x1 0' 5 Q cm and 1.5x1 O' 6 Q cm at 300 K and 77 K, respectively. Although the absolute values were increased by roughly 10 4 orders of magnitude, the resistivity curve indicates that the (La,Sr)Ti0 3 film remains metallic until the liquid nitrogen temperature of 77 K. The resistivity results of post annealed sample at YBa 2 Cu 3 0 7 deposition condition show that

PAGE 88

73 the (La,Sr)Ti03 layer can be a candidate for the conductive buffer layer in the coated conductor applications. 4.4 Conclusions The perovskite (La,Sr)Ti03 was investigated as a possible conducting oxide buffer layer for high temperature superconducting coated conductors. In order to observe the oxygen sensitivity, thin (La,Sr)Ti03 films were epitaxially deposited by PLD on single crystal SrTiC>3 substrates at various oxygen partial pressures and compared with LaTiC>3 films at the same growth conditions. The room temperature resistivity of LaTi03 increases rapidly as the oxygen pressure increases by more than two orders of magnitude. In contrast, the resistivity of (La,Sr)Ti03 films is relatively insensitive to oxygen pressure making it attractive as a conductive buffer for coated conductor applications. The high temperature superconducting layer such as YBa2Cu307 was grown epitaxially on (La,Sr)Ti03 buffer layer on SrTiC^ substrate with excellent in-plane and out-of-plane alignment. The superconducting transition temperature (T c ) of YBa2Cu307 / (La,Sr)Ti03 / SrTi03 structure was 91 K and the critical current density (J c ) of this structure was 2.18 x 10 6 A/cm 2 at 0 magnetic field. The resistivity results of post annealed sample at YBa2Cu3C>7 deposition condition indicates that the (La,Sr)Ti03 layer can be a candidate for the conductive buffer layer in the coated conductor applications.

PAGE 89

Intensity (arbitrary units) 74 40 60 20 (degrees) Figure 4-1. The X-ray diffraction 0-20 scan of (La,Sr)Ti03 film grown on SrTiC>3 single crystal substrate by pulsed laser deposition (PLD) method at 750°C, in vacuum.

PAGE 90

75 Figure 4-2. The in situ reflection high-energy electron diffraction (RHEED) pattern of (La,Sr)Ti 03 film deposited on SrTiC >3 single crystal substrate by PLD method at 750°C, in vacuum. RHEED is used to monitor the epitaxial film growth.

PAGE 91

Intensity (arbitrary units) 76 20 (degrees) Figure 4-3. The X-ray 0-20 scans of (La,Sr)Ti 03 films deposited on SrTiC >3 single crystal substrates in different ambient conditions.

PAGE 92

Intensity (arbitrary units) 77 10 7 10 6 10 5 10 4 10 3 10 2 10 * 10 ° 20 40 60 80 20 (degrees) Figure 4-4. The X-ray diffraction 9-20 scan of LaTiC >3 film grown on SrTi 03 single crystal substrate by pulsed laser deposition (PLD) method at 750°C, in vacuum.

PAGE 93

Intensity (arbitrary units) 78 20 (degrees) Figure 4-5. The two different X-ray 0-20 scans of LaTi 03 film grown on SrTi03 single crystal substrate in the oxygen pressure of 3.0x1 O' 5 Torr, aligned to (a) the substrate (001) plane and (b) the La 2 Ti 207 (-210) plane, respectively.

PAGE 94

Intensity (arbitrary units) 79 10 4 10 ° 10' 4 20 40 60 20 (degrees) 80 Figure 4-6. The two different X-ray 0-20 scans of LaTiC >3 film grown on SrTiC >3 single crystal substrate in the oxygen pressure of 4.0x1 O' 4 Torr, aligned to (a) the substrate (00 1 ) plane and (b) the La 2 Ti 207 (-2 1 0) plane, respectively.

PAGE 95

Intensity (arbitrary units) 80 10 3 10 2 0 90 180 270 360 (j> (degrees) Figure 4-7. The X-ray (j)-scan of (-420) peak for La2Ti20 7 layer grown on SrTiC >3 single crystal substrate in the oxygen pressure of 3.0x1 0' 5 Torr.

PAGE 96

Resistivity (nQ cm) Resistivity (^Q cm) 81 0 2 Pressure (Torr) Figure 4-8. The resistivities of (La,Sr)Ti03 and LaTiC>3 films on single crystal SrTiC>3, measured at (a) 300 K and (b) 77 K, as a function of the oxygen pressure.

PAGE 97

82 Temperature (K) Figure 4-9. The resistivity curves of (La,Sr)Ti03 films on SrTi03 substrates grown in (a) vacuum, (b) 3.0x1 O' 5 Torr of oxygen, and (c) 4.0x1 O' 4 Torr of oxygen, as a function of temperature.

PAGE 98

83 Figure 4-10. The resistivity curves of LaTiC >3 films on SrTi 03 substrates grown in (a) vacuum, (b) 3.0x1 OÂ’ 5 Torr of oxygen, and (c) 4.0x1 OÂ’ 4 Torr of oxygen, as a function of temperature.

PAGE 99

84 20 (degrees) Figure 4-1 1. The X-ray 0-20 scan of YBa2Cu307 deposited on (La,Sr)Ti03 buffer layer on SrTi03 single crystal substrate.

PAGE 100

85 0 (degrees) Figure 4-12. The X-ray co-scan of (a) (002) peak for (La,Sr)TiC >3 buffer layer on SrTi 03 single crystal substrate and (b) (006) peak for the YBa 2 Cu 307 film deposited on (La,Sr)Ti 03 buffer layer.

PAGE 101

Intensity (arbitrary units) Intensity (arbitrary units) 86 10 5 10 4 10 3 10 2 10 4 10 3 10 2 -180 -90 0 90 180 4> (degrees) Figure 4-13. The X-ray 3 buffer layer on SrTiC>3 single crystal substrate and (b) (012) peak for the YBa2Cu307 film deposited on (La,Sr)Ti03 buffer layer.

PAGE 102

87 Temperature (K) Figure 4-14. The resistivity versus temperature measurement for YBa2Cu307 film deposited on (La,Sr)Ti03 buffer layer on single crystal SrTiC>3 substrate.

PAGE 103

88 H (T) Figure 4-15. The critical current density (Jc) as a function of magnetic field of YBa 2 Cu 307 film deposited on (La,Sr)TiC >3 buffer layer on single crystal SrTiC >3 substrate.

PAGE 104

89 Temperature (K) Figure 4-16. The resistivity versus temperature graph of (La,Sr)Ti03 film grown on single crystal SrTiC>3 substrate after annealing with the YBa2Cu307 deposition condition.

PAGE 105

CHAPTER 5 EPITAXIAL (LA,SR)TI0 3 AS A CONDUCTIVE BUFFER FORNI-W BASED HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 5.1 Introduction LaTi0 3+x is a quite attractive material for coated conductor application because it has good metallic property at all temperature range in the oxygen stoichiometric of 0. 1 < x < 0.25 [137], and its pseudo-cubic lattice parameter is well matched with YBa 2 Cu 3 07 . However, the composition also can be changed during YBa 2 Cu 3 07 growth in the high oxygen ambient (~1 xlO' 1 Torr) at high temperature (~780°C). The previous research on the relation of resistivity of LaTi0 3+x with oxygen pressure during deposition by PLD indicated that oxygen played a crucial role in conducting property [139], Cation doping in the compound can overcome this oxygen sensitivity of LaTi0 3+x . Doping with divalent element increases the Ti +3 / Ti 4+ ratio and can make the compound less sensitive to ambient oxygen pressure during YBCO deposition. Electrical conductivity is a function of doping constant x in the Lai. x Sr x Ti0 3 compound, and the resistivity continuously decreases with higher cation doping [140], Previous study on oxygen dependency showed the room temperature resistivity of Lafl. 5 Sro. 5 Ti0 3 was remained below 1.0x1 O' 3 Qcm in 10 -4 ~ 10' 2 Torr of oxygen pressure range [141], The metallic (La,Sr)Ti0 3 film is being considered for various electronic applications [168-170]. including coated onductors based on epitaxial high temperature superconducting (HTS) films deposited on metal tapes. In functioning as a conducting buffer layer in a RABiTS structure, the conductive oxide layer architecture must satisfy a specific set of criteria. First, the buffer layer must 90

PAGE 106

91 be reasonably well lattice matched to both the metal substrate and the superconducting film, thus enabling epitaxy. Second, the material must be conductive not only as deposited, but after subsequent HTS film growth and oxygen annealing. Third, the interaction between the buffer layer and the metal substrate must be such as to minimize formation of any native interfacial oxide that would serve as an insulating barrier to shunted current flow. (La,Sr)Ti 03 appears to in satisfy these criteria. At room temperature, LaTiC >3 possesses the orthorhombic GdFeC >3 perovskite structure with a = 5.604 A, b = 5.595 A, and c = 7.906 A. The pseudo-cubic lattice parameter of 3.96 A provides a relatively large (12%) lattice mismatch to Ni (a = 3.524 A), although similar mismatched parameters have proven useful in other RABiTS architectures. La, Sr, and Ti have a high affinity for oxygen relative to Ni, suggesting that native NiO formation at the interface should be minimal. In order to grow conductive (La,Sr)TiC >3 layer on Ni-W based metal tape, TiN was selected as a seed layer for templating the epitaxial film growth. Recent research has suggested that considerable sharpening of the out-of-plane texture occurs in YBa 2 Cu 307 _s films grown on oxide buffer layers due to a tilted epitaxy growth mechanism in the TiN seed [167], The buffer scheme using transition metal nitride film was reported by Kim et al. in 2002 [69]. In this report, TiN was chosen because of low electrical resistivity (20-30 pQcm) and good mechanical strength (YoungÂ’s modulus : 600 GPa, microhardness : 2000 Kg/mm 2 ). Due to the oxidation of TiN layer during YBCO deposition, CeC >2 layer also applied between YBCO and TiN films. The TiN layer was formed by DC reactive sputtering with Ar/N 2 mixture gas and the Ce02, YBCO layers were grown by PLD method. A superconducting transition temperature for this architecture was 89 K

PAGE 107

92 and the critical current density was 6xl0 5 A/cm 2 at 77 K. Because TiN is also known as Cu diffusion barrier [70], Cantoni et al. studied multi buffer scheme using TiN for Cu based RABiTS process [71]. In this report, LaMnC>3 / MgO / TiN buffer architecture was proposed. MgO layer was chosen for oxygen diffusion barrier, and LaMn03 layer proved to be a planarizing material for smooth growing of YBCO. 5.2 Experiments The Ni-3% W alloy tapes were used as a metal substrate. The Ni-W tapes have { 100}<100> cube texture and obtained from randomly oriented metal bars by coldrolling, followed by an anneal in vacuum at 800°C for 1 hour. The growth of (La,Sr)Ti03 films was performed by pulsed laser deposition (PLD) in vacuum at 750°C, at an energy of 2 J/cm 2 and repetition rate of 5~10Hz with a thickness of 0.6pm. A KrF (248nm) excimer laser was used as the ablation source. For the Ni-W based RABiTS application, PLD TiN layer was selected as a (La,Sr)Ti03 template. The deposition condition for TiN was in 1 .Ox 1 O' 4 Torr of nitrogen at 650°C, energy density of 3.5 J/cm 2 , rate 1 0Hz, with a thickness range of 0.05-0. 5pm. YBa2Cu307_s layer was grown as a high temperature superconductor by PLD under conditions of 1 ,0x 10’ 1 Torr of oxygen at 780°C, energy density of 2 J/cm 2 , 10Hz rate, with a thickness of 0.2pm. After completing YBCO deposition, the PLD chamber was cooled down to 500°C at a rate of 28°C/min and the oxygen pressure was increased to 400 Torr. After 20 minutes, the chamber was cooled down to room temperature rapidly. The x-ray diffraction of 0-20 scan, co and <|>-scan were used in order to observe the in-plane and out-of-plane alignments of each layers. The in situ reflection high-energy electron diffraction (RHEED) was used to monitor epitaxial film growth. The surface

PAGE 108

93 observation and the compositional analysis of thin films were investigated by scanning electron microscope (SEM) and energy dispersive x-ray spectroscopy (EDS). A standard four probe technique was used to evaluate the electrical properties, including superconducting transition temperature ( T c ) and the critical current density (J c ). The J c measurements were carried out in liquid nitrogen, defined at a 1 pV/cm criterion. The widths of the samples were in the range of 0.40 ~0.50cm and the distance between the voltage tips was in the range of 0.3~0.4cm. 5.3 Results and Discussion The X-ray diffraction 0-20 scan shown in Figure 5-1 is for (La,Sr)Ti 03 films grown directly on Ni-3% W alloy tape. Although the Ni-W tape has { 1 00}<1 00> cube texture, (La,Sr)Ti 03 layers grown at the temperature range of 700°C ~ 800°C showed strong (111) peaks. In order to grow (La,Sr)TiC >3 layer epitaxially on Ni-W tape without (111) peaks, the transition metal nitride such as TiN was selected for a epitaxial seed layer. The proper thickness of TiN layer should be considered because the metal elements can be diffused through TiN layer and reach the sample surface. The metal atoms diffused out from the metal substrate can lower the transition temperature of YBa 2 Cu 307 . In order to adjust the appropriate thickness, the TiN films with the thickness range of 500A ~ 5000A were deposited on 10000A thick Cu films which were grown on single crystal SrTi 03 substrates and the samples were annealed at 740°C in vacuum for 30 minutes and 60 minute. Figure 5-2 shows the energy dispersive X-ray spectroscopy (EDS) of the samples with different thicknesses after annealing for 60 minutes. The EDS signals were captured from the surfaces of TiN films. According to figure 5-2, titanium (Ti) and nitrogen (N) peaks were very weak in the sample of 500A thick TiN layer and gradually increased as

PAGE 109

94 the TiN layer were getting thicker. This indicates that the level of metal diffusion from the substrate can be affected by the TiN layer thickness. Figure 5-3 is the atomic percent of Cu observed on the surface of the TiN layer. The Cu composition level decreases drastically as the TiN thickness increases. The sample of 500A thick TiN layer showed clear difference of Cu composition before and after annealing. From the annealing test, the proper TiN thickness was thought to be around 2000A because thicker film has higher tensile stress which generates the handling problem. Figure 5-4 is the X-ray diffraction 020 scans of TiN layer deposited on Ni-W tape and (La,Sr)TiC>3 film grown on TiN seed layer by PLD. The (La,Sr)Ti03 (1 12) peaks observed in the sample which was grown directly on the Ni-W tape vanished and (00/) peaks were clearly defined. Figure 5-5 shows the in situ monitored reflection high-energy electron diffraction (RHEED) patterns of TiN layer and (La,Sr)Ti03 film grown on TiN seed layer. According to the x-ray 0-20 scan and RHEED pattern, the TiN film is epitaxially deposited on textured Ni-W alloy tape. In addition to that, the (La,Sr)TiC>3 buffer layer also formed epitaxially on the TiN seed layer. From the X-ray diffraction to-scans shown in figure 5-6, rocking curve sharpening was observed. The full width half maximum (FWHM) value of (002) TiN showed 3.4° compared to (002) Ni-W tape of 8.7°. FWHM value for the TiN film is considerably smaller than the corresponding value for the Ni-W substrate. FWHM of (004) (La,Sr)Ti03 grown on TiN layer showed 3.8°. The in-plane aligned textures of the films are shown in the X-ray (j)-scans through the Ni (111), TiN (111) and (La,Sr)Ti0 3 (1 12). The large peaks every 90 degrees in figure 5-7 represent the majority in-plane aligned (La,Sr)Ti03 and TiN grains. The FWHM values of the <|>-scan are 7.28°, 6.48°, 8.89° for

PAGE 110

95 Ni (1 1 1), TiN (1 1 1), and (La,Sr)TiC>3 (1 12), respectively. Figure 5-7 indicates that a cube-on-cube epitaxial relationship was obtained among the Ni-W substrate, TiN and (La,Sr)Ti03 films. According to the x-ray diffraction 0-20 scan, co-scan, ())-scan and RHEED study, in-plane and out-of-plane aligned epitaxial thin film of (La,Sr)Ti03 was grown on Ni-W metal tape by using TiN as a seed layer. In order to assess the compatibility of (La,Sr)Ti03 as a buffer layer for RABiTS application, the growth of YBa 2 Cu 3 0 7 _8 thin film on (La,Sr)Ti0 3 / TiN / Ni-W tape structure was investigated. The X-ray diffraction 0-20 scan in figure 5-8 shows clear YBa 2 Cu30 7 -g peaks. Although the YBa 2 Cu30 7 .g thin film was grown under conditions of 1.0x1 O' 1 Torr of oxygen at 780°C, neither Ni oxide nor Ti oxide peaks detected in the scan. Figure 5-9 is the x-ray diffraction co-scan of (005) YBa 2 Cu30 7 _g and figure 5-10 is the (|)-scan of (103) YBa 2 Cu30 7 _§ on (La,Sr)Ti03 buffer layer / TiN seed layer / Ni-W tape structure. The rocking curve FWHM value of the (005) YBa 2 Cu 3 0 7 _ 6 peak is 4.92° and scan FWHM value of (103) YBa 2 Cu30 7 .g peak is 10.1°. Both x-ray diffraction results suggest that the in-plane and out-of-plane aligned high temperature superconducting YBa 2 Cu30 7 _g film can be formed epitaxially on the (La,Sr)TiC>3 / TiN / Ni-W tape structure. A standard four-probe technique was used to evaluate the electrical properties, including superconducting transition temperature ( T c ) and critical current density (J c ). The J c value was assigned at a 1 pV/cm criterion. Figure 5-11 shows the temperaturedependent net resistivity of the YBa 2 Cu 3 0 7 _§ on (La,Sr)Ti0 3 / TiN / Ni-W tape. The sample has a T c value of 89 K which is lower than the standard YBa 2 Cu30 7 _§ film. Figure 5-12 is the I-V curve obtained from the YBa 2 Cu30 7 _s on (La,Sr)Ti03 / TiN / Ni-W tape in

PAGE 111

96 a field of 0 T and at the temperature of 77 K. The zero-field transport J c is 0.42x1 0 6 A/cm . Figure 5-13 is the SEM picture of YBa2Cu307_s film surface grown on (La,Sr)Ti03 / TiN / Ni-W tape which shows YBa2Cu307^ 2 nd phases and pin holes. These could cause the lower T c and ,/ c values and deviation of resistivity versus temperature curve from the ideal behavior. Figure 5-14 is the EDS analysis graph indicating that the YBa2Cu307-s film does not possess any cation contamination. 5.4 Conclusions The epitaxial film growth of (La,Sr)Ti03 was examined on Ni-W metal alloy tape. The transition metal nitride such as TiN was deposited epitaxially on Ni-W tape by PLD and played an excellent role as a seed layer for (La,Sr)TiC>3 film growth on Ni-W tape. The high temperature superconducting YBa2Cu3C>7 film was deposited epitaxially on (La,Sr)Ti03 buffer layer with the TiN seed layer on Ni-W tape. The YBa2Cu3C>7 film grown on (La,Sr)Ti03 / TiN / Ni-W tape has superconducting transition temperature of 89 K and critical current value of 0.42x1 0 6 A/cm 2 . This shows that (La,Sr)Ti0 3 is a possible candidate for the conductive buffer layer in the coated conductor applications.

PAGE 112

97 20 (degrees) Figure 5-1. The X-ray diffraction 9-20 scan of (La,Sr)Ti03 films deposited directly on Ni-W tape by PLD at the temperature of (a) 700°C, (b) 750°C, and (c) 800°C.

PAGE 113

Intensity (arbitrary units) 98 Figure 5-2. The energy dispersive X-ray spectroscopy (EDS) results of TiN films grown on Cu layers on single crystal SrTiCE substrates with the thicknesses of (a) 500A, (b) 1000A, (c) 2000A, and (d) 5000A. These curves were taken after annealing the samples at 740°C, in vacuum for 60 minutes.

PAGE 114

99 Figure 5-3. The atomic percent of Cu observed on the surface of TiN layer with the thickness range of 500~5000A. The square symbols designate the samples without annealing. The circles and triangles show the samples with annealing at 740°C, in vacuum for 30 minutes and 60 minutes, respectively.

PAGE 115

Intensity (arbitrary units) Intensity (arbitrary units) 100 20 (degrees) 20 (degrees) Figure 5-4. The X-ray diffraction 0-20 scan of (a) TiN layer deposited on textured Ni-W alloy tape, and (b) (La,Sr)Ti03 film deposited on TiN seed layer.

PAGE 116

101 Figure 5-5. The in situ reflection high-energy electron diffraction (RHEED) pattern of (a) TiN film deposited on textured Ni-W alloy tape, and (b) (La,Sr)Ti03 film deposited on TiN seed layer.

PAGE 117

102 15 20 25 30 35 0 (degrees) Figure 5-6. The X-ray diffraction co-scan of (a) Ni-W (002), (b) TiN (002), and (c) (La,Sr)Ti 03 (004) planes.

PAGE 118

103 <|> (degrees) Figure 5-7. The X-ray diffraction <(>-scan of (a) Ni-W (111), (b) TiN (1 1 1), and (c) (La,Sr)Ti03 (112) planes.

PAGE 119

104 20 (degrees) Figure 5-8. The X-ray diffraction 0-20 scan of high temperature superconducting YBa2Cu3C>7 film grown on (La,Sr)Ti03 buffer layer / TiN seed layer / Ni-W tape.

PAGE 120

105 Figure 5-9. The X-ray diffraction c)-scan of (005) YBa 2 Cu 3 C >7 which was grown on (La,Sr)Ti0 3 buffer layer / TiN seed layer / Ni-W tape.

PAGE 121

Intensity (arbitrary units) 106 -180 -90 0 90 180 <|) (degrees) Figure 5-10. The X-ray diffraction (|>-scan of (103) YBa2Cu307 which was grown on (La,Sr)Ti03 buffer layer / TiN seed layer / Ni-W tape.

PAGE 122

107 Temperature (K) Figure 5-11. The resistivity versus temperature graph for YBa 2 Cu 30 7 film grown on (La,Sr)Ti 03 buffer layer / TiN seed layer / Ni-W tape.

PAGE 123

(A) A 108 1(A) Figure 5-12. The voltage versus current graph of YBa 2 Cu 3 0 7 film deposited on (La,Sr)TiC >3 buffer layer / TiN seed layer / Ni-W tape.

PAGE 124

109 Figure 5-13. The SEM picture of YBa2Cu307 film surface deposited on (La,Sr)Ti03 buffer layer / TiN seed layer / Ni-W tape.

PAGE 125

110 Energy (K eV) Figure 5-14. The energy dispersive X-ray spectroscopy (EDS) results of YBa2Cu 3 0 7 film grown on (La,Sr)Ti0 3 / TiN / Ni-W tape.

PAGE 126

CHAPTER 6 EPITAXIAL (LA,SR)TI0 3 AS A CONDUCTIVE BUFFER FOR CU BASED HIGH TEMPERATURE SUPERCONDUCTING COATED CONDUCTORS 6.1 Introduction The important factor that can influence the current density, J c in the superconducting wire application is the crystallinity of the HTS material. The high-angle grain boundaries generated in the polycrystalline HTS can reduce the amount of current because the grain boundary can act as weak superconducting interface separating the superconductivity. This is known as weak links and the grain boundary effect was demonstrated by Chaudhari et al. [34], The YBaaCusO? film grown on textured metal templates can drastically reduce the misorientation of the individual grains allowing improvement of the links in the current path. The metal tapes produced by the thermomechanical texturing is known as rolling assisted biaxially textured substrate (RABiTS). The BABiTS process also consists of depositing buffer layers and HTS materials on the biaxially textured flexible metal substrates. Nickel which is the current starting template of 2G wires has ferromagnetic (FM) property and the magnetic metal substrate such as Ni can cause significant hysteretic losses during application of alternating current (ac). The ac loss can be decreased by adding W to Ni [78]. However, even the Ni-W (5 at.%) alloy tape is not perfect nonmagnetic. In this aspect, Cu substrate is another candidate for the starting template of 2G wires. Cu is a diamagnetic material which shows no ac loss phenomenon. Cu also surpasses Ni alloy in the electrical conductivity characteristics. At 300 K, the resistivity 111

PAGE 127

112 of Cu tape is 1.5x1 O' 6 Q-cm which is lower than Ni-W (3 at.%) alloy tape of 2.5x1 O' 5 Qcm. At the temperature of liquid nitrogen 77 K, Cu tape and Ni-W (3 at.%) tape show the resistivity value of 2. Oxl O' 7 Q-cm and 1.7x1 O' 5 Q-cm respectively. Cu tape has { 100}<100> cube texture and obtained from randomly oriented metal bars by coldrolling, followed by an anneal in vacuum at 800°C for 1 h. Though the Cu based RABiTS process has advantages, the Cu oxidation problem could be worse. The high temperature and high oxygen ambient of YBa 2 Cu 307 growth condition can cause copper oxidation. One method of preventing Cu oxidation is applying noble metal as an oxygen barrier between metal substrate and conductive oxide buffer layers. One of the candidate material is iridium (Ir) which has face centered cubic (fee) crystal structure and lattice parameter of 3.840A. In the crystallographic aspect, iridium can be well matched with Cu substrate. Because iridium is well known platinum group metal with excellent oxidation resistance, oxidation behavior is only considered above 1400°C [95~97], Although the iridium oxides (I 1 O 2 ) are formed after the YBCO deposition, it has still good metallic property with the resistivity of below 3xl0‘ 4 Q-cm at room temperature [98], Aytug et al. reported Ir as an oxygen barrier on Ni-W alloy based RABiTS application [99]. In their research, Lao 7 Sr 0 3 Mn 03 conductive buffer layer was used to form fully conductive buffer architecture. On the other hand, careful consideration is also needed, because oxygen diffusivity through Ir is 5x1 0‘ 12 cm 2 /s at 800°C [78] and there is a possibility of Cu diffusion through Ir layer. Cu and Ir are known to be soluble in the extremely small amount of atomic % [100]. However, 50pm thick Cu substrate can continuously supply copper element to the thin Ir layer during buffer oxides and YBCO processes. These mean that Cu oxidation can be observed either

PAGE 128

113 on top of the YBCO film or on the Cu substrate and proper thickness of Ir layer should be deposited. In this chapter, novel buffer scheme for high temperature superconducting coated conductors will be addressed. Multi layer of (La,Sr)Ti03 and Ir will be formed on Cu tape. The metallic (La,Sr)Ti03 serves as a template for YBa2Cu307 film growth and Ir layer acts as an oxygen diffusion barrier between the metal substrate and the buffer layer. The (La,Sr)Ti03 and Ir stack will be the fully conductive buffer scheme on Cu based RABiTS application. 6.2 Experiments The copper tapes were used as a metal substrate. The Cu tapes have { 100}<100> cube texture and obtained from randomly oriented metal bars by cold-rolling, followed by an anneal in vacuum at 800°C for 1 hour. The growth of (La,Sr)Ti03 films was performed by pulsed laser deposition (PLD) in vacuum at 750°C, at an energy of 2 J/cm 2 and repetition rate of 5~10Hz with a thickness of 0.6p.m. A KrF (248nm) excimer laser was used as the ablation source. Ir seed layers were used as an oxygen barrier before (La,Sr)TiC>3 deposition on Cu tapes. Thin films of Ir were grown by PLD in 7.0x1 O’ 4 Torr of 4% FL/Ar mixture gas at 300°C, energy density of 4 J/cm 2 , 40Hz, with thickness below 0. 1 pm. YBajCusCL^ layer was grown as a high temperature superconductor by PLD under conditions of 1.0x1 O' 1 Torr of oxygen at 780°C, energy density of 2 J/cm 2 , 1 0Hz rate, with a thickness of 0.2pm. After completing YBCO deposition, the PLD chamber was cooled down to 500°C at a rate of 28°C/min and the oxygen pressure was increased to 400 Torr. After 20 minutes, the chamber was cooled down to room temperature rapidly.

PAGE 129

114 The x-ray diffraction of 0-20 scan, co and <)>-scan were used in order to observe the in-plane and out-of-plane alignments of each layers. The in situ reflection high-energy electron diffraction (RHEED) was used to monitor epitaxial film growth. The surface observation and the compositional analysis of thin films were investigated by scanning electron microscope (SEM) and energy dispersive x-ray spectroscopy (EDS). A standard four probe technique was used to evaluate the electrical properties, including superconducting transition temperature ( T c ) and the critical current density ( J c ). The J c measurements were carried out in liquid nitrogen, defined at a 1 pV/cm criterion. The widths of the samples were in the range of 0.40 ~0.5 0cm and the distance between the voltage tips was in the range of 0.3~0.4cm. 6.3 Results and Discussion One of the clue that the Cu substrate has a critical problem of oxidation is showing in the figure 6-1 and figure 6-2. After depositing (La,Sr)Ti03 film on Cu tape, annealing was done at 780°C in oxygen 1.0x1 O' 1 Torr for 7 minutes which is the YBa2Cu307 film growth condition in PLD system. According to these SEM and EDS figures, the surface of the as deposited film of (La,Sr)Ti03 was covered with the Cu oxide. For this reason, iridium was selected as an oxygen diffusion barrier on Cu tape. Figure 6-3 shows the Xray diffraction 0-20 scan of Ir film deposited on textured Cu substrate and (La,Sr)Ti03 film grown on Ir layer. Figure 6-4 shows the in situ monitored reflection high-energy electron diffraction (RHEED) patterns of Ir layer deposited on Cu tape and (La,Sr)TiC>3 film grown on Ir layer. According to the x-ray 0-20 scan and RHEED pattern, the Ir film is epitaxially deposited on textured Cu tape. In addition to that, the (La,Sr)TiC>3 buffer layer also formed epitaxially on the Ir layer. From the X-ray diffraction w-scans shown in

PAGE 130

115 figure 6-5, the full width half maximum (FWHM) value of (002) Ir showed 6.20° which is close to the substrate Cu (002) of 5.41°. (La,Sr)TiC >3 film grown on Ir showed FWHM value of 12.3°. The in-plane aligned texture of the (La,Sr)TiC >3 film which is grown on Ir layer is shown in the X-ray <|>-scan. The large peaks every 90 degrees in figure 6-6 represent the majority in-plane aligned (La,Sr)TiC> 3 . The FWHM value of the (f>-scan is 8.63°. Figure 6-7 shows the pole figure graphs of (1 1 1) Cu, (1 1 1) Ir and (112) (La,Sr)Ti 03 . The pole figure is generated by scanning X-ray signals with rotating the sample 360° (3 film to Ir layer, temperature dependent resistivity measurement was performed. Figure 6-8 shows the resistivity curves of Ir film, (La,Sr)TiC >3 film grown on Ir layer and (La,Sr)TiC >3 film, respectively. All the films were grown on SrTiC >3 single crystal substrates. The (La,Sr)TiC >3 films of these samples were deposited at 612°C, in vacuum by PLD system. The resistivity curve of (La,Sr)TiC >3 film deposited on Ir layer shows quite different shape from the (La,Sr)TiC >3 film itself. And it follows the Ir curve shape even showing lower values. This indicates that the (La,Sr)TiC >3 film is electrically well connected to the underlying Ir layer. Figure 6-9 designates the resistivity curves of Ir films grown on single crystal SrTiC >3 substrate before and after annealing at 780°C in oxygen pressure of

PAGE 131

116 l.OxlO' 1 Torr for 7 minutes. In this figure, Ir shows no resistivity change even after annealing at the severe oxidizing condition. In order to assess the compatibility of (La,Sr)Ti03 as a buffer layer for Cu based RABiTS application, the growth of YBa2Cu307^ thin film on (La,Sr)Ti03 / Ir / Cu tape structure was investigated. The X-ray diffraction 0-20 scan in figure 6-10 shows clear YBa2Cu307_a peaks. Because the YBa2Cu307_s thin film was grown under conditions of l.OxlO' 1 Torr of oxygen at 780°C, copper oxide peaks were observed in this figure. Figure 6-1 1 shows the defects on the surface of the YBa2Cu3C>7 film of the sample. Figure 6-12 is the energy dispersive X-ray spectroscopy graphs of the normal YBa2Cu3C>7 film surface and the defect region of the YBa2Cu3C>7 film. The normal surface of the film shows only Y, Ba, Cu, O contents. On the other hand, the defect region shows Cu and O peaks which indicate that the defects are copper oxides. These copper oxides were formed by the reaction of oxygen gas and the Cu elements which were diffused out from the Cu tape. A standard four-probe technique was used to evaluate the electrical properties, including superconducting transition temperature (T c ) and critical current density (J c ). The J c value was assigned at a 1 pV/cm criterion. Figure 6-13 shows the temperaturedependent net resistivity of the YBa2Cu3()7_g on (La,Sr)Ti03 and Ir buffer stack on Cu tape. The sample has a T c value of 90 K and the zero-field transport J c was 1 .Ox 1 0 6 A/cm . These results indicate that the metallic (La,Sr)TiC>3 layer played an excellent role of templating HTS film. And fully conductive buffer layer scheme could be obtained in the Cu based RABiTS applications. According to the X-ray 0-20 scan, surface SEM pictures and the EDS analysis of the YBa2Cu3C>7 film, the thicknesses of (La,Sr)TiC>3 and

PAGE 132

117 Ir buffers should be considered carefully to prevent forming copper oxide either on top of YBaaCuaOy or in the interface between the Cu tape and the buffer layers. 6.4 Conclusions The epitaxial film growth of (La,Sr)Ti 03 was examined on Cu tape as a possible conducting buffer layer for high temperature superconducting coated conductors. The noble metal such as Ir was deposited epitaxially on Cu tape by PLD for oxygen diffusion barrier. The high temperature superconducting YBa 2 Cu 307 film was deposited epitaxially on (La,Sr)Ti 03 and Ir buffer stack on Cu tape. The YBa 2 Cu 3 C >7 film grown on (La,Sr)TiC >3 / Ir / Cu tape has superconducting transition temperature of 90 K and critical current density value of 1 .OxlO 6 A/cm 2 . This shows that (La,Sr)Ti 03 is a possible candidate for the conductive buffer layer in the Cu based RABiTS applications.

PAGE 133

118 Figure 6-1. The surface SEM picture of the (La,Sr)Ti 03 film grown on Cu tape after annealing at 780°C in oxygen partial pressure of 1 .0x1 O' 1 Torr for 7 minutes.

PAGE 134

Intensity (arbitrary units) 119 Energy (K eV) Figure 6-2. The energy dispersive X-ray spectroscopy (EDS) graph of (La,Sr)Ti 03 film grown on Cu tape after annealing at 780°C in oxygen 1.0x1 O' 1 Torr for 7 minutes.

PAGE 135

Intensity (arbitrary units) Intensity (arbitrary units) 120 20 (degrees) 20 (degrees) Figure 6-3. The X-ray diffraction 0-20 scan of (a) Ir layer deposited on textured Cu by PLD, and (b) (La,Sr)TiC >3 film deposited on Ir layer.

PAGE 136

121 Figure 6-4. The in situ reflection high-energy electron diffraction (RHEED) pattern of Ir film deposited on textured Cu tape, and (b) (La,Sr)Ti03 film deposited on Ir layer.

PAGE 137

122 V) "5 3 i— IS 3 s 3 b 3 15 •— 3 t fi 3 a > 1000 3 b 3 2 -*•* 15 La 3 (Z3 C OJ 15 20 25 30 0 (degrees) Figure 6-5. The X-ray diffraction co-scan of (a) Cu (002), (b) Ir (002), and (c) (La,Sr)Ti 03 (004) planes.

PAGE 138

123 0 90 180 270 360 <|> (degrees) Figure 6-6. The X-ray diffraction 3 (1 12) which was grown on Ir film on Cu tape.

PAGE 139

124 HI a ^ 1 7. 1 N 173 IB volume fraction : 89 . 13 % :: volume fraction : 85 . 38 % volume fraction : 88 . 65 % 262 1.11 -17 Figure 6-7. The X-ray pole figures of (a) Cu (1 1 1), (b) Ir (1 1 1), and (c) (La,Sr)Ti 03 ( 112 ).

PAGE 140

125 Figure 6-8. The resistivity curves of Ir on SrTiC>3 single crystal, (La,Sr)Ti03 film on SrTi03 substrate and (La,Sr)Ti03 film on Ir on SrTi03 substrate.

PAGE 141

126 Figure 6-9. The resistivity curves of Ir on SrTiC >3 single crystal with and without annealing at 780°C in oxygen 1.0x1 O' 4 Torr for 7 minutes.

PAGE 142

Intensity (arbitrary units) 127 20 (degrees) Figure 6-10. The X-ray diffraction 0-20 scan of YBa 2 Cu 307 layer deposited on (La,Sr)Ti 03 / Ir multi buffer stack on textured Cu tape.

PAGE 143

128 Figure 6-11. The surface SEM picture of the YBa 2 Cu 307 film grown on (La,Sr)Ti 03 / Ir buffer stack on Cu tape with the magnification of (a) xlOOO and (b) x5000.

PAGE 144

129 Q \m lP.li i i d\vm . .hm > 0 200 400 600 800 1000 Energy (eV) Figure 6 12 . The energy dispersive X-ray spectroscopy (EDS) graph of (a) normal YBa2Cu3C>7 film surface, (b) defect region of YBa2Cu307 film.

PAGE 145

Resistivity (jifi*cm) 130 Figure 6-13. The resistivity versus temperature measurement for YBa2Cu3C>7 film deposited on (La,Sr)Ti03 / Ir multi buffer stack on Cu tape.

PAGE 146

CHAPTER 7 SUMMARY Epitaxial growth of intermetallic Cu 2 Mg was investigate for the Cu based RABiTS applications. Epitaxial growth of (La,Sr)Ti 03 layer was investigate for the conductive buffer architecture of Cu based high temperature superconducting coated conductors. In order to form the oxidation resistant buffers for RABiTS-based conductors, the epitaxial growth of (Cu,Mg) films on a (001) Cu surface was investigated. The approach used to achieve epitaxy of (Cu,Mg) films was to form Cu/Mg multilayer precursor films that would be subsequently annealed to form either Mg-doped fee Cu or intermetallic Cu 2 Mg. The precursor consists of an Mg/Cu multilayer stack with 5 each of 25 nm thick Mg and 25 nm thick Cu layers which were grown at room temperature by sputter deposition. Annealing was performed in a flowing H 2 /Ar mixture. At annealing temperature of 400°C, formation of the intermetallic Cu 2 Mg was observed. X-ray diffraction rocking curve for the (004) Cu 2 Mg peak showed a full-width half-maximum (FWHM) of A0 = 2.0°, which is slightly larger than that for the Cu film (A0 = 1 .45°). An in-plane (p-scan through the Cu 2 Mg (222) indicates that the grains are epitaxial with respect to the Cu film, possessing a cube-on-cube orientation. The FWHM of the in-plane peak is 2.1°. In order to confirm the oxidation resistance of the structures possessing (Cu,Mg) alloy films, CeC >2 films were deposited at elevated temperature on Ni/(Cu,Mg)/Cu/MgO and on Ni/Cu/MgO. In case of the CeC >2 film on Ni/Cu/MgO, significant surface roughness due to the metal oxidation is observed. In contrast, no 131

PAGE 147

132 surface roughness is observed in the SEM image for the Ce02/Ni/(Cu,Mg)/Cu/Mg0 structure. The perovskite (La,Sr)Ti03 was investigated as a possible conducting oxide buffer layer for high temperature superconducting coated conductors. In order to observe the oxygen sensitivity, thin (La,Sr)Ti03 films were epitaxially deposited by PLD on single crystal SrTi03 substrates at various oxygen partial pressures and compared with LaTiC>3 films at the same growth conditions. The room temperature resistivity of LaTi03 increases rapidly as the oxygen pressure increases by more than two orders of magnitude. In contrast, the resistivity of (La,Sr)Ti03 films is relatively insensitive to oxygen pressure making it attractive as a conductive buffer for coated conductor applications. The high temperature superconducting layer such as YBaiC^Oy was grown epitaxially on (La,Sr)Ti03 buffer layer on SrTiC>3 substrate with excellent in-plane and out-of-plane alignment. The superconducting transition temperature (T c ) of YBa2Cu3(>7 / (La,Sr)Ti03 / SrTi03 structure was 91 K and the critical current density (J c ) of this structure was 2.18*10 6 A/cm 2 at 0 magnetic field. The resistivity results of post annealed sample at YBa2Cu3C>7 deposition condition indicates that the (La,Sr)Ti03 layer can be a candidate for the conductive buffer layer in the coated conductor applications. The epitaxial film growth of (La,Sr)Ti03 was examined on Ni-W metal alloy tape. The transition metal nitride such as TiN was deposited epitaxially on Ni-W tape by PLD and played an excellent role as a seed layer for (La,Sr)TiC>3 film growth on Ni-W tape. The high temperature superconducting YBa2Cu3C>7 film was deposited epitaxially on (La,Sr)Ti03 buffer layer with the TiN seed layer on Ni-W tape. The YBa2Cu3C>7 film

PAGE 148

133 grown on (La,Sr)Ti03 / TiN / Ni-W tape has superconducting transition temperature of 89 K and critical current value of 0.42x 1 0 A/cm . The epitaxial film growth of (La,Sr)TiC>3 was examined on Cu tape as a possible conducting buffer layer for high temperature superconducting coated conductors. The noble metal such as Ir was deposited epitaxially on Cu tape by PLD for oxygen diffusion barrier. The high temperature superconducting YBa2Cu3C>7 film was deposited epitaxially on (La,Sr)TiC>3 and Ir buffer stack on Cu tape. The YBa2Cu3C>7 film grown on (La,Sr)Ti03 / Ir / Cu tape has superconducting transition temperature of 90 K and critical current density value of 1 .Ox 1 0 A/cm .

PAGE 149

LIST OF REFERENCES 1. M.R. Beasley, and T.H. Geballe, Phys. Today 36, 60 (1984) 2. D.C. Johnston, H. Prakash, W.H. Zachariasen, and R. Viswanathan, Mat. Res. Bull. 8, 777(1973) 3. T.D. Thanh, A. Koma, and S. Tanaka, Appl. Phys. 22, 205 (1980) 4. J. Bardeen, L.N. Cooper, and J.R. SchriefFer, Physical Review 108, 1 175 (1957) 5. P. Townsend, and J. Sutteon, Physical Review 128, 591 (1962) 6. Frank J. Blatt, Modem Physics, McGraw-Hill (1992) 7. F. London, and H. London, Proc. Roy. Soc. (London) 149, 71 (1935) 8. G. Bums, High-Temperature Superconductivity an Introduction, Academic Press, Inc., San Diego (1992) 9. C.P. Poole, Jr., H.A. Farach, and R.J. Creswick, Superconductivity, Academic Press, San Diego (1995) 1 0. J.G. Bednorz and K.A. Muller, Z. Phys. B Condensed Matter 64, 1 89 (1986) 11. M.K. Wu, J.R. Ashbum, C.J. Tomg, P.H. Hor, R.L. Meng, L.Gao, Z.J. Huang, Y.Q. Wang, and C.W. Chu, Physical Review Letters 58, 908 (1987) 12. H. Maeda, Y. Tanaka, M. Fukutomi, and T. Asano, Japanese Journal of Applied Physics 27, L209(1988) 13. Z.Z. Sheng, W. Kiehl, J. Bennett, A. El Ali, D. Marsh, G.D. Mooney, F. Arammash, J. Smith, D. Viar, and A.M. Hermann, Appl. Phys. Lett. 52, 1738 (1988) 14. S.N. Putilin, E.V. Antipov, and M. Marezio, Physica C, 212, 266 (1993) 15. D.P. Norton, Annu. Rev. Mater. Sci. 28, 299 (1998) 134

PAGE 150

135 16. M.A. Beno, L. Soderholm, D.W. Capone, II, D.G. Hinks, J.D. Jorgensen, J.D. Grace, Ivan K. Schuller, C.U. Segre, and K. Zhang, Appl. Phys. Lett. 51, 57 (1987) 17. P.K. Gallagher, H.M. OÂ’Bryan, S.A. Sunshine, and D.W. Murphy, Mat. Res. Bull. 22, 995 (1987) 18. R.J. Cava, B. Batlogg, C.H. Chen, E.A. Rietman, S.M. Zahurak, and D. Werder, Nature 329, 423 (1987) 19. A. Santoro, S. Miraglia, and F. Beech, Mat. Res. Bull. 22, 1007 (1987) 20. P.M. Mankiewich, J.H. Scofield, W.J. Skocpol, R.E. Howard, A.H. Dayem, and E. Good, Appl. Phys. Lett. 51, 1753 (1987) 21. R. Feenstra, L.A. Boatner, J.D. Budai, D.K. Christen, M.D. Galloway, and D.B. Poker, Appl. Phys. Lett. 54, 1063 (1989) 22. Y. Zhang, R. Feenstra, J.R. Thompson, A.A. Gapud, T. Aytug, P.M. Martin, and D.K. Christen, Supercond. Sci. Technol. 17, 1154 (2004) 23. C.B. Eom, J.Z. Sun, K. Yamamoto, A.F. Marshall, K.E. Luther, T.H. Geballe, and S.S. Laderman, Appl. Phys. Lett. 55, 595 (1989) 24. D.P. Norton, D.H. Lowndes, X.-Y. Zheng, S. Zhu, and R.J. Warmack, Physical Review B 44, 9760(1991) 25. J. Akimitsu, A. Yamazaki, H. Sawa, and H. Fujiki, Japanese Journal of Applied Physics 26, L2080(1987) 26. C. Michel, M. Hervieu, M.M. Borel, A. Grandin, F. Deslandes, J. Provost, and B. Raveau, Z. Phys. B Condensed Matter 68, 421 (1987) 27. S.A. Sunshine, T. Siegrist, L.F. Schneemeyer, D.W. Murphy, R.J. Cava, B. Batlogg, R.B. van Dover, R.M. Fleming, S.H. Glarum, S. Nakahara, R. Farrow, J.J. Krajewski, S.M. Zahurak, J.V. Waszczak, J.H. Marshall, P. Marsh, L.W. Rupp, Jr., and W.F. Peck, Physical Review B 38, 893 (1988) 28. M. Kanai, T. Kawai, and S. Kawai, Appl. Phys. Lett. 54, 1802 (1989) 29. M. Kanai, T. kawai, M. Kawai, and S. Kawai, Japanese Journal of Applied Physics 27, LI 293 (1988) 30. M.A. Subramanian, C.C. Torardi, J.C. Calabrese, J. Gopalakrishnan, K.J. Morrissey, T.R. Askew, R.B. Flippen, U. Chowdhry, and A.W. Sleight, Science 239, 1015 (1988)

PAGE 151

136 31. H. Tabata, T. Kawai, M. Kanai, O. Murata, and S. Kawai, Japanese Journal of Applied Physics 28, L430 (1989) 32. T. Kawai, Y. Egami, H. Tabata, and S. Kawai, Nature 349, 200 (1991) 33. G.B. Lubkin, Physics Today 49, 48 (1996) 34. P. Chaudhari, J. Mannhart, D. Dimos, C.C. Tsuei, J. Chi, M.M. Oprysko, and M. Scheuermann, Physical Review Letters 60, 1653 (1988) 35. A. Goyal, D.P. Norton, J.D. Budai, M. Paranthaman, E.D. Specht, D.M. Kroeger, D.K. Christen, Q. He, B. Saffian, F.A. List, D.F. Lee, P.M. Martin, C.E. Klabunde, E. Hartfield, and V.K. Sikka, Appl. Phys. Lett. 69, 1795 (1996) 36. A. Goyal, D.P. Norton, D.K. Christen, E.D. Specht, M. Paranthaman, D.M. Kroeger, J.D. Budai, Q. He, F.A. List, R. Feenstra, H.R. Kerchner, D.F. Lee, E. Hatfield, P.M. Martin, J. Mathis, and C. Park, Applied Superconductivity 4, 403 (1996) 37. Selected Powder Diffraction Data for Metals and Alloys, First edition. Volume I, JCPDS-Intemational Centre for Diffraction Data, Pennsylvania (1978) 38. E.D. Specht, A. Goyal, D.F. Lee, F.A. List, D.M. Kroeger, M. Paranthaman, R.K. Williams, and D.K. Christen, Supercond. Sci. Technol. 11, 945 (1998) 39. A. Goyal, R. Feenstra, M. Paranthaman, J.R. Thompson, B.Y. Kang, C. Cantoni, D.F. Lee, F.A. List, P.M. Martin, E. Lara-Curzio, C. Stevens, D.M. Kroeger, M. Kowalewski, E.D. Specht, T. Aytug, S. Sathyamurthy, R.K. Williams, and R.E. Ericson, Physica C 382, 25 1 (2002) 40. A.O. Ijaduola, J.R. Thompson, A. Goyal, C.L.H. Thieme, and K. Marken, Physica C 403, 163 (2004) 41. J.R. Thompson, A. Goyal, D.K. Christen, and D.M. Kroeger, Physica C 370, 169 ( 2002 ) 42. A. Goyal, M. Parans Paranthaman, and U. Schoop, MRS Bulletin 29, 552 (2004) 43. T.J. Jackson, B.A. Glowacki, and J.E. Evetts, Physica C 296, 215 (1998) 44. J.M.S. Skakle, Materials Science and Engineering R23, 1 (1998) 45. G. Xiao, F.H. Streitz, A. Gavrin, Y.W. Du, and C.L. Chien, Physical Review B 35, 8782 (1987)

PAGE 152

137 46. Y. Maeno, T. Tomita, M. Kyogoku, S. Awaji, Y. Aoki, K. Hoshino, A. Minami, and T. Fujita, Nature 328 , 512 (1987) 47. P. Strobel, C. Paulsen, and J.L. Tholence, Solid State Communications 65 , 585 (1988) 48. J.M. Tarascon, P. Barboux, P.F. Miceli, L.H. Greene, and G.W. Hull, Physical Review B 37, 7458 (1998) 49. E. T-Muromachi, Y. Uchida, and K.Kato, Japanese Journal of Applied Physics 26 , L2087 (1987) 50. S. Witanachi, S. Patel, H.S. Kwok, and D.T. Shaw, Appl. Phys. Lett. 54 , 578 (1989) 51. A.B. Berezin, C.W. Yuan, and A.L. de Lozanne, Appl. Phys. Lett. 57 , 90 (1990) 52. Y. Hakuraku, K. Maezono, and H. Ueda, Supercond. Sci. Technol. 9, 775 (1996) 53. K. Char, M.S. Colclough, S.M Garrison, N. Newman, and G. Zaharchuk, Appl. Phys. Lett. 59 , 733 (1991) 54. B.K. Moon, and H. Ishiwara, Jpn. J. Appl. Phys. 33, 1472 (1994) 55. H. Ishiwara, N. Tsuji, H. Mori, and H. Nohira, Appl. Phys. Lett. 61, 1459 (1992) 56. M.G. Blamire, G.W. Morris, R.E. Somekh, and J.E. Evetts, J. Phys. D: Appl. Phys. 20 , 1330(1987) 57. K. Hirata, K. Yamamoto, K. Iijima, J. Takada, T. Terashima, Y. Bando, and H. Mazaki, Appl. Phys. Lett. 56 , 683 (1990) 58. M. Migliuolo, A.K. Stamper, D.W. Greve, and T.E. Schlesinger, Appl. Phys. Lett. 54 , 859(1989) 59. H. Myoren, Y. Nishiyama, N. Miyamoto, Y. Kai, Y. Yamanaka, Y. Osaka, and F. Nishiyama, Japanese Journal of Applied Physics 29 , L955 (1990) 60. Q.Z. Jia, S.Y. Lee, W.A. Anderson, and D.T. Shaw, Appl. Phys. Lett. 59, 1120 (1991) 61. M.W. Denhoff, and J.P. McCaffrey, J. Appl. Phys. 70, 3986 (1991) 62. T. Venkatesan, E.W. Chase, X.D. Wu, A. Inam, C.C. Chang, and F.K. Shokoohi, Appl. Phys. Lett. 53 , 243 (1988)

PAGE 153

138 63. K. Harada, N. Fujimori, and S. Yazu, Japanese Journal of Applied Physics 27, L1524 (1988) 64. L.S. Hung, J.A. Agostinelli, G.R. Paz-Pujalt, and J.M. Mir, Appl. Phys. Lett. 53 , 2450(1988) 65. D.P. Norton, A. Goyal, J.D. Budai, D.K. Christen, D.M. Kroeger, E.D. Specht, Q. He, B. Saffian, M. Paranthaman, C.E. Klabunde, D.F. Lee, B.C. Sales, and F.A. List, Science 274 , 755 (1996) 66. M. Paranthaman, A. Goyal, F.A. List, E.D. Specht, D.F. Lee, P.M. Martin, Q. He, D.K. Christen, D.P. Norton, J.D. Budai, and D.M. Kroeger, Physica C 275 , 266 (1997) 67. Q. He, D.K. Christen, J.D. Budai, E.D. Specht, D.F. Lee, A. Goyal, D.P. Norton, M. Paranthaman, F.A. List, and D.M. Kroeger, Physica C 275 , 155 (1997) 68. J.E. Mathis, A. Goyal, D.F. Lee, F.A. List, M. Paranthaman, D.K. Christen, E.D. Specht, D.M. Kroeger, and P.M. Martin, Jpn. J. Appl. Phys. 37, LI 379 (1998) 69. I. Kim, P.N. Barnes, A. Goyal, S.A. Barnett, R. Biggers, G. Kozlowski, C. Varanasi, I. Maartens, R. Nekkanti, T. Peterson, T. Haugan, and S. Sambasivan, Physica C 377, 227 (2002) 70. S-Q. Wang, I. Raaijmakers, B.J. Burrow, S. Suthar, S. Redkar, and K-B. Kim, J. Appl. Phys. 68,5176(1990) 71. C. Cantoni, D.K. Christen, M. Varela, J.R. Thompson, S.J. Pennycook, E.D. Specht, and A. Goyal, J. Mater. Res. 18 , 2387 (2003) 72. L.J. Masur, D. Buczek, E. Harley, T. Kodenkandath, X. Li, J. Lynch, N. Nguyen, M. Rupich, U. Schoop, J. Scudiere, E. Siegal, C. Thieme, D. Verebelyi, W. Zhang, and J. Kellers, Physica C 392 396 , 989 (2003) 73. H. Maeda, Y. Tanaka, M. Fukutomi, T. Asano, K. Togano, H. Kumakura, M. Uehara, S. Ikeda, K. Ogawa, S. Horiuchi, and Y. Matsui, Physica C 153 155 , 602 (1988) 74. K. Heine, J. Tenbrink, and M. Thoner, Appl. Phys. Lett. 55, 2441 (1989) 75. M. Parans Paranthaman, and T. Izumi, MRS Bulletin 29, 533 (2004) 76. M.W. Rupich, U. Schoop, D.T. Verebelyi, C. Thieme, W. Zhang, X. Li, T. Kodenkandath, N. Nguyen, E. Siegal, D. Buczek, J. Lynch, M. Jowett, E. Thompson, J-S. Wang, J. Scudiere, A.P. Malozemoff, Q. Li, S. Annavarapu, S.

PAGE 154

139 Cui, L. Fritzemeier, B. Aldrich, C. Craven, F. Niu, R. Schwall, A. Goyal, and M. Paranthaman, IEEE Trans. Appl. Supercond. 13 , 2458 (2003) 77. R.E. Flummel, Electronic Properties of Materials, Springer-Verlag, New York ( 2001 ) 78. A. Goyal, M. Parans Paranthaman, and U. Schoop, MRS Bulletin 29 552 (2004) 79. A.O. Ijaduola, J.R. Thompson, A. Goyal, C.L.H. Thieme, and K. Marken, Physica C 403 , 163 (2004) 80. C. Cantoni, T. Aytug, D.T. Verebelyi, M. Paranthaman, E.D. Specht, D.P. Norton, and D.K. Christen, IEEE Transactions on Applied Superconductivity 11, 3309(2001) 81 . E.G. Brentari and R.V. Smith, Adv. Cryog. Eng. 10 , 325 (1965) 82. S. Pinol, J. Diaz, M. Segarra, and F. Espiell, Supercond. Sci. Technol. 14 , 1 1 ( 2001 ) 83. T. Aytug, B.W. Kang, C. Cantoni, E.D. Specht, M. Paranthaman, A. Goyal, D.K. Christen, D.T. Verebelyi, J.Z. Wu, R.E. Ericson, C.L. Thomas, C-Y. Yang, and S.E. Babcock, Journal of Materials Research 16 , 2661 (2001) 84. X.D. Wu, S.R. Foltyn, R.C. Dye, Y. Coulter, and R.E. Muenchausen, Appl. Phys. Lett. 62 , 2434(1993) 85. K. Sreedhar, J.M. Honig, M. Darwin, M. McElfresh, P.M. Shand, J. Xu, B.C. Crooker, and J. Spalek, Physical Review B 46 , 6382 (1992) 86. T. Aytug, M. Paranthaman, J.R. Thompson, A. Goyal, N. Rutter, FLY. Zhai, A.A. Gapud, A.O. Ijaduola, and D.K. Christen, Applied Physics Letters 83 , 3963 (2003) 87. H.M. Naguib, R.J. Kriegler, J.A. Davies, and J.B. Mitchell, J. Vac. Sci. Techno. 13,396(1976) 88. W.A. Lanford, P.J. Ding, W. Wang, S. Hymes, and S.P. Muraka, Thin Solid Films 262 , 234(1995) 89. Z-C. Wu, Y-L. Liu, and M-C. Chen, Thin Solid Films 346, 186 (1999) 90. Z-C. Wu, Y-L. Liu, and M-C. Chen, Thin Solid Films 358, 1 80 (2000) X.Q. Zhao, Y.F. Han, and B.X. Liu, J. Appl. Phys. 90, 1638 (2001) 91.

PAGE 155

140 92. B. Arcot, S.P. Murarka, L.A. Clevenger, Q.Z. Hong, W. Ziegler, and J.M.E. Harper, J. Appl. Phys. 76, 5161 (1994) 93. M. Hansen and K. Anderko, Constitution of Binary Alloys, McGraw-Hill Book, New York (1958) 94. W.H. Lee, H.L. Cho, B.S. Cho, J.Y. Kim, W.J. Nam, Y-S. Kim, W.G. Jung, H. Kwon, J.H. Lee, and J.G. Lee, Applied Physics Letters 77, 2191 (2000) 95. H. Jehn, Journal of the Less-Common Metals 100, 321 (1984) 96. R.T. Wimber and H.G. Kraus, Metallurgical Transactions 5, 1565 (1974) 97. H. Jehn, R. Volker, and M.I. Ismail, Platinum Met. Rev. 22, 92 (1978) 98. M.A.E. Khakani and M. Chaker, Thin Solid Films 335, 6 (1998) 99. T. Aytug, M. Paranthaman, H.T. Zhai, A.A. Gapud, K.J. Leonard, P.M. Martin, A. Goyal, J.R. Thompson, and D.K. Christen, Applied Physics Letters 85, 2887 (2004) 100. W.G. Moffatt, The Handbook of Binary Phase Diagrams, Genium Publishing Corp., New York (1984) 101. M. Ohring, Materials Science of Thin Films, Academic Press, San Diego (2002) 102. H.S. Kim, K.C. Chung, B.S. Lee, S.M. Lim, and D. Youm, Supercond. Sci. Technol. 17, 291 (2004) 103. G. Yuan, J. Yang, and K. Shi, IEEE Trans, on Appl. Supercond. 11, 3382 (2001) 104. M.P. Paranthaman, T. Aytug, S. Kang, R. Feenstra, J.D. Budai, D.K. Christen, P.N. Arendt, L. Stan, J.R. Groves, R.F. DePaula, S.R. Foltyn, and T.G. Holesinger, IEEE Trans, on Appl. Supercond. 13, 2481 (2003) 105. S.A. Campbell, The Science and Engineering of Microelectronic Fabrication, Oxford University Press, New York (1996) 106. R. Metzger, M. Bauer, K. Numssen, R. Semerad, P. Berberich, and H. Kinder, IEEE Trans, on Appl. Supercond. 11, 2826 (2001) 107. L.D. Chang, M.Z. Tseng, and E.L. Hu, Appl. Phys. Lett. 60, 1753 (1992) 108. A. Ichinose, C-Y. Yang, D.C. Larbalestier, S.E. Babcock, A. Kikuchi, K. Tachikawa, and S. Akita, Physica C 324, 1 13 (1999)

PAGE 156

141 109. A.P. Malozemoff, S. Annavarapu, L. Fritzemeier, Q. Li, V. Prunier, M. Rupich, C. Thieme, W. Zhang, A. Goyal, M. Paranthaman, and D.F. Lee, Supercond. Sci. Technol. 13 , 473 (2000) 1 1 0. D.B. Chrisey and G.K. Hubler, Pulsed Laser Deposition of Thin Films, John Wiley & Sons, Inc., New York (1994) 111. H. Iwai, T. Muroga, T. Watanabe, S. Miyata, Y. Yamada, Y. Shiohara, T. Kato, and T. Hirayama, Supercond. Sci. Technol. 17, 496 (2004) 1 12. C. Park, D.P. Norton, D.K. Christen, D.T. Verebelyi, R. Feenstra, J.D. Budai, A. Goyal, D.F. Lee, E.D. Specht, D.M. Kroeger, and M. Paranthaman, IEEE Trans, on Appl. Supercond. 9, 2276 (1999) 1 13. C. Cantoni, D.K. Christen, E.D. Specht, M. Varela, J.R. Thompson, A. Goyal, C. Thieme, Y. Xu, and S.J. Pennycook, Supercond. Sci. Technol. 17 , 341 (2004) 1 14. A. Usoskin and H.C. Freyhardt, MRS Bulletin 29, 583 (2004) 115. R.W. Schwartz, Chem. Mater. 9, 2325 (1997) 116. M. Paranthaman, S.S. Shoup, D.B. Beach, R.K. Williams, and E.D. Specht, MRS Bulletin 32, 1697 (1997) 117. S. Sathyamurthy, M. Paranthaman, H-Y. Zhai, H.M. Christen, P.M. Martin, and A. Goyal, J. Mater. Res. 17, 2181 (2002) 1 1 8. M.S. Bhuiyan, M. Paranthaman, S. Sathyamurthy, T. Aytug, S. Kang, D.F. Lee, A. Goyal, E.A. Payzant, and K. Salama, Supercond. Sci. Technol. 16 , 1305 (2003) 119. P.N. Arendt and S.R. Foltyn, MRS Bulletin 29, 543 (2004) 120. J. Zaanen, G.A. Sawatzky, and J.W. Allen, Physical Review Letters 55, 41 8 (1985) 121 . J.B. Torrance, P. Lacorre, C. Asavaroengchai, and R.M. Metzger, Physica C 182, 351 (1991) 122. R.D. Sanchez, M.T. Causa, A. Caneiro, A. Butera, M. Vallet-Regi, M.J. Sayagues, and J. Gonzalez-Calbet, Physical Review B 54, 16574 (1996) 123. Q. He, D.K. Christen, R. Feenstra, D.P. Norton, M. Paranthaman, E.D. Specht, D.F. Lee, A. Goyal, and D.M. Kroeger, Physica C 314 , 105 (1999) 124. A. Callaghan, C.W. Moeller, and R. Ward, Inorganic Chemistry 5, 1572 (1966)

PAGE 157

142 125. L. Klein, J.S. Dodge, C.H.Ahn, J.W. Reiner, L. Mieville, T.H. Geballe, M.R. Beasley, and A. Kapitulnik, J. Phys.:Condens. Matter 8, 10111 (1996) 126. P. Tiwari, X.D. Wu, S.R. Foltyn, M.Q.Le, I.H. Campbell, R.C. Dye, and R.E. Muenchausen, Appl. Phys. Lett. 64 , 634 (1994) 127. T. Aytug, J.Z. Wu, C. Cantoni, D.T. Verebelyi, E.D. Specht, M. Paranthaman, D.P. Norton, D.K. Christen, R.E. Ericson, and C.L. Thomas, Applied Physics Letters 76, 760 (2000) 128. C. Cantoni, T. Aytug, D.T. Verbelyi, M. Paranthaman, E.D. Specht, D.P. Norton, and D.K. Christen, IEEE Trans, on Appl. Supercond. 11 , 3309 (2001) 129. Q.X. Jia, S.R. Foltyn, P.N. Arendt, T. Holesinger, J.R. Groves, and M. Hawley, IEEE Trans, on Appl. Supercond. 13 , 2655 (2003) 130. J.M.D. Coey, M. Viret, and S. von Molnar, Advances in Physics 48 , 167 (1999) 131. C. Zener, Physical Review 82, 403 (1951) 132. D.P. Norton, Materials Science and Engineering R 43 , 139 (2004) 133. T. Aytug, M. Paranthaman, B.W. Kang, S. Sathyamurthy, A. Goyal, and D.K. Christen, Applied Physics Letters 79, 2205 (2001) 134. Y. Okada, T. Arima, Y. Tokura, C. Murayama, and N. Mori, Physical Review B 48 , 9677 (1993) 135. A. Schmehl, F. Lichtenberg, H. Bielefeldt, J. Mannhart, and D.G. Schlom, Applied Physics Letters 82 , 3077 (2003) 136. K. Yoshii, A. Nakamura, and H. Abe, J. of Alloys and Compounds 290 , 236 (1999) 137. F. Lichtenberg, D. Widmer, J.G. Bednorz, T. Williams, and A. Reller, Z. Phys. B-Condensed Matter 82, 21 1 (1991) 138. J.G. Bednorz, Physica C 282-287, 37 (1997) 139. K.H. Kim, D.P. Norton, J.D. Budai, M.F. Chisholm, B.C. Sales, D.K. Christen, and C. Cantoni, phy. stat. sol. (a) 200 , 346 (2003) 140. C.C. Hays, J.-S. Zhou, J.T. Markert, and J.B. Goodenough, Physical Review B 60 , 10367(1999)

PAGE 158

143 141. K. Kim, Y.W. Kwon, D.P. Norton, D.K. Christen, J.D. Budai, B.C. Sales, M.F. Chisholm, C. Cantoni, and K. Marken, Solid-State Electronics 47, 2177 (2003) 142. K. Kim, D.P. Norton, C. Cantoni, T. Aytug, A. A. Gapud, M. Paranthaman, A. Goyal, and D.K. Christen, IEEE Trans, on Appl. Supercond., to be published (2005) 143. G. Prokhorov, G.G. Kaminsky, I.I. Kravchenko, and Y.P. Lee, Physica B 324, 205 (2002) 144. G. Yalovega and A.V. Soldatov, phys. stat. sol. (b) 218, 455 (2000) 145. H.C. Yu and K.Z. Fung, J. Mater. Res. 19, 943 (2004) 146. K.M. Satyalakshmi, R.M. Mallya, K.V. Ramanathan, X.D. Wu, B. Brainard, D.C. Gautier, N.Y. Vasanthachary, and M.S. Hegde, Appl. Phys. Lett. 62, 1233 (1993) 147. X.Q. Xu, J.L. Peng, Z.Y. Li, H.L. Ju, and R.L. Greene, Physical Review B 48, 1112(1993) 148. C.N.R. Rao and B. Raveau, Transition Metal Oxides, John Wiley & Sons, Inc., New York (1998) 149. A.J. Millis, P.B. Littlewood, and B.I. Shraiman, Physical Review Letters 74, 5144(1995) 150. G. Demazeau, C. Parent, M. Pouchard, and P. Hagenmuller, Mat. Res. Bull. 7, 913 (1972) 151. D.P. Norton, C. Park, C. Prouteau, D.K. Christen, M.F. Chisholm, J.D. Budai, S.J. Pennycook, A. Goyal, E.Y. Sun, D.F. Lee, D.M. Kroeger, E. Specht, M. Paranthaman, and N.D. Browning, Materials Science and Engineering B56, 86 (1998) 152. S. Mrowec, and A. Stoklosa, Oxidation of Metal 3, 291 (1971) 153. M. Hansen, and K. Anderko, Constitution of Binary Alloys, McGraw-Hill Book Company, New York, 594 (1958) 1 54. F.A. Shunk, Constitution of Binary Alloys, Second Supplement, McGraw-Hill Book Company, New York, 290 (1969) 155. K. Schroder, Handbook of Electrical Resistivities of Binary Metallic Alloys, CRC Press Inc., Boca Raton, FL, 219 (1983)

PAGE 159

144 156. D. Goo, S. Oh, J. Kim, K. Jung, and D. Youm, Physica C 305, 221 (1998) 157. T.J. Jackson, B.A. Glowacki, and J.E. Evetts, Physica C 296, 215 (1998) 158. R.P. Wang, Y.L. Zhou, S.H. Pan, M. He, H.B. Lu, Z.H. Chen, G.Z. Yang, C.F. Liu, X. Wu, F.Y. Wang, Y. Feng, P.X. Zhang, X.Z. Wu, and L. Zhou, Physica C 337, 87 (2000) 159. X.D. Wu, S.R. Foltyn, P.N. Arendt, W.R. Blumenthal, I.H. Campbell, J.D. Cotton, J.Y. Coulter, W.L. Hults, M.P. Maley, H.F. Safar, and J.L. Smith, Appl. Phys. Lett. 67, 2397(1995) 160. G. Sipos, B. Utz, W. Schmidt, H.W. Neumueller, and P. Mueller, Physica C 341, 1457(2000) 161. M. Jin, S.C. Han, T.H. Sung, and K. No, Physica C 334, 243 (2000) 162. S.B. Kim, T. Maeda, Y. Yamada, T. Suga, Y. Yamada, T. Watanabe, K. Matsumoto, and I. Hirabayashi, Physica C 357-360, 979 (2001) 163. K. Matsumoto, S.B. Kim, K. Yamagiwa, Y. Koike, I. Hirabayashi, T. Watanabe, N. Uno, and M. Ikeda, Physica C 335, 39 (2000) 164. G. Celentano, C. Annino, V. Boffa, L. Ciontea, F. Fabbri, U. Gambardella, V. Galluzzi, G. Grimaldi, A. Mancini, L. Muzzi, and T. Petrisor, Physica C 341-348, 2501 (2000) 165. S. Gariglio, J.W. Seo, J. Fompeyrine, J.-P. Locquet, and J.-M. Triscone, Physical Review B 63, 161 103(R) (2001) 166. D.P. Norton, K. Kim, D.K. Christen, J.D. Budai, B.C. Sales, M.F. Chisholm, D.M. Kroeger, A. Goyal, and C. Cantoni, Physica C 372-376, 818 (2002) 167. C. Cantoni, A. Goyal, U. Schoop, X. Li, M.W. Rupich, C. Thieme, A. A. Gapud, T. Aytug, M. Paranthaman, K. Kim, J.D. Budai, and D.K. Christen, IEEE Trans, on Appl. Supercond., to be published (2005) 168. B.T. Liu, K. Maki, Y. So, V. Nagarajan, R. Ramesh, J. Lettieri, J.H. Haeni, D.G. Schlom, W. Tian, X.Q. Pan, F.J. Walker, and R.A. Mckee, Appl. Phys. Lett. 80, 4801 (2002) 169. T. Hato, A. Yoshida, C. Yoshida, H. Suzuki, and N. Yokoyama, Appl. Phys. Lett. 70, 2900(1997) 1 70. G.Q. Li, P.T. Lai, S.H. Zeng, M.Q. Huang, and Y.C. Cheng, Sensors and Actuators A 63, 223 (1997)

PAGE 160

BIOGRAPHICAL SKETCH Kyunghoon Kim was bom in Seoul, Republic of Korea, in November, 1964. He grew up and was educated in his hometown. Kyunghoon received a B.S. degree in ceramics engineering from Yonsei University in Seoul, Korea, in 1988. After graduation he entered the R&D center of Samsung Semiconductor Inc. in Kiheung, Korea. He participated in the project of developing capacitor process for DRAM devices. From 1988 to 1992, he developed silicon oxide and silicon nitride films for capacitor dielectrics. From 1992 to 1994, he developed hemi spherical grain silicon and various storage node structures for capacitance enhancement. From 1994 to 1999, he developed a high dielectric film process for a 1 giga bit DRAM device. The high dielectric film process includes the Ta 20 s and AI 2 O 3 as high dielectric films and TiN film as an electrode material. He continued his study after work time at night and obtained his masterÂ’s degree in 1996. The research topic for his masterÂ’s degree was a study on the fabrication and characteristics of the tantalum oxide capacitor according to the nitrdation pre-treatment. In 2000, he was admitted to the Department of Materials Science and Engineering at the University of Florida to pursue his Ph.D degree with a specialty in electronic materials. His research area at Dr. NortonÂ’s group was synthesis and characterization of conductive oxides on metal substrate. Furthermore, he extended his research interests into the application of high temperature superconducting coated conductor by collaboration in the Oak Ridge National Laboratory since 2002. 145

PAGE 161

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in j^ope/md quality, as a dissertation for the degree of Doctor of Philosopl; David P. Norton, Chair Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. y A Cammy R. Abema^Ky Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Rajiv Singh Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Wolfgang'Sigmund Associate Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and qjiality, as a dissertation for the degree of Doctor of Philosophy. Andrew Rinzler Associate Professor of Physics

PAGE 162

This dissertation was submitted to the Graduate Faculty of the College of Engineering and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 2005 1 Pramod P. Khargonekar Dean, College of Engineering Kenneth Gerhardt Interim Dean, Graduate School