l/

/MICHIGAN snr

I ll

I////II///I//////I///I/Il ill/limit ‘

9195

 

 

" fflfc':

'This is to certify that the

thesis entitled

IN SITU LASER ABLATION DEPOSITION OF YBaZCu307_x THIN

FILMS USING A MICROWAVE PLASMA DISK REACTOR OXYGEN SOURCE

presented by
Conrad Matthew Pawlowski

has been accepted towards fulfillment
of the requirements for

h7‘ S ‘ degree in E6

 

/£L ~ flab/x

Major professor

Datej’jZ?/C22/

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

LiBHARY
Michigan State
L University

 

PLACE iN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

   
 

DATE DUE DATE DUE DATE DUE

JUN 1 8m
:9 ‘

       

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

MSU is An Affirmative Action/Equal Opportunity institution

«me“:

 

 

 

 

 

IN SIT U LASER ABLATION DEPOSITION OF YBazCu3O7,x THIN FILMS USING
A MICROWAVE PLASMA DISK REACTOR OXYGEN SOURCE

By

Conrad Matthew Pawlowski

A THESIS

Submitted to Michigan State University in partial fulfillment of the requirements for
the degree of

MASTER OF SCIENCE

Department of Electrical Engineering

1992

 

 

aflé

/

70/ K 4/")

 

IN SIT U LASER ABLATION DEPOSITION OF YBazCu3O7,x FILMS USING A
MICROWAVE PLASMA DISK REACTOR OXYGEN SOURCE

Conrad Matthew Pawlowski

In situ laser ablation deposition of high-temperature superconductors (HTSC) has
been employed to produce device-quality films for possible applications in microsensors,
Josephson junctions, VLSI interconnects and vacuum microelectronics. Understanding
the effects of lowered substrate temperatures (<650 °C) and lowered oxygen partial
pressures (<40 mTorr) is important to improving film quality. Under these conditions,
molecular oxygen is insufficient for the necessary formation of the orthorhombic-II
superconducting phase (X=6.93) of YBazCu3Ox. This phase has recently been grown at
lowered pressures by laser ablation with atomic oxygen. Utilizing electron cyclotron
resonance (ECR) magnets, a microwave plasma disk reactor (MPDR) can produce high
concentrations of atomic oxygen (>IO”/cm3) at low pressures (<10 mTorr). The goals of
this research were to design a laser ablation system, optimize YBCO growth by laser
ablation and to use atomic oxygen from an MPDR oxygen source for low-pressure (<10
mTorr) deposition. We successfully deposited YBCO films using the MPDR at low
oxygen partial pressures (4 mTorr, 815 °C). Films exhibited zero resistance at
temperatures as high as 79 K and the measured critical current was as high as 2.9 x 10’

A/cm2 at 77 K.

 

Copyright by

CONRAD MATTHEW PAWLOWSKI

1992

ACKNOWLEDGEMENTS

The author wishes to acknowledge and thank Dr. M. Aslam of Michigan State
University and Dr. L. Rimai of the Ford Motor Company research staff for guidance,
support, and communication throughout this project. Support from the Center for
Fundamental Materials Research made this project possible. Support was also received
from Ford Motor Company which provided equipment to accomplish these studies. At
various stages of this project, support in the form of communication was also helpful from
D. Kubinski, D. Hoffman, E. M. Logothetis, and R. Soltis. R. Ager generated X-ray
diffraction patterns and R. Elder helped with SEM images. Thanks must also be extended

to A. Srivastava and J. Asmussen for communication and assistance regarding the

operation of the MPDR-610.

TABLE OF CONTENTS

LIST OF TABLES ............................................ vii
LIST OF FIGURES .......................................... viii
Introductuction ........................................... 1
Chapter 1
Basics of Superconductivity
1.1 Low-Temperature Superconductors and Superconducting Phenomena ...... 4
1.1.1 Zero-Resistance State ................................. 4
1.1.2 Meissner Effect ..................................... 4
1.1.3 Josephson Effect .................................... 7
1.1.4 Flux-quantization Effect .............................. 10
1.2 High-Temperature Superconductors ............................ 10
1.2.1 Physical Properties .................................. 11
1.2.2 Crystal Structure ................................... 12
1.3 Theories of Superconductivity ............................... 13
1.3.1 Electrical PrOperties ................................. 13
1.3.2 BCS Theory for Low-Temperature Superconductors ........... 14
1.3.3 RVB Theory and Goddard's Magnon Theory for HTSC ......... 16
Chapter 2
YBCO Thin-Film Deposition
2. 1 Introduction ............................................ l 7
2.2 Laser Ablation, Sputtering and Electron-Beam Evaporation ............ 17
2.3 Theoretical Model Laser Ablation ............................. 19
2.4 Novel YBCO Thin Film Deposition Techniques ................... 24
2.4.1 Novel Laser Ablation Techniques ....................... 24
2.4.2 Activated Oxygen Enriched Film Deposition ................ 25

3.1
3.2
3.3
3.4

4.1
4.2

4.3

5.4
5.5
5.6

6.1
6.2

Chapter 3
YBCO Laser Ablation Deposition Parameters

Substrates ............................................. 27
Target Conditions ........................................ 29
Laser Types and Conditions ................................. 30
Oxygen Partial Pressure Effects .............................. 32

Chapter 4

Design and Construction of the YBCO Deposition System

Introduction ............................................ 34
Design Considerations ..................................... 34
4.2.1 Chamber Design .................................... 34
4.2.2 Microwave Plasma Disk Reactor ........................ 35
4.2.3 Optics ........................................... 39
4.2.4 Target Port ....................................... 39
4.2.5 Vacuum Pump ..................................... 39
4.2.6 Substrate Heater ................................... 40
YBCO Growth Process Design Considerations .................... 42
4.3.1 Target—To-Substrate Distance ........................... 42
4.3.2 Length of Deposition ................................ 43
4.3.3 Cooling Procedure .................................. 43

Chapter 5

Results

Introduction ............................................ 46
Experimental Procedure for YBCO Depositions ................... 47
Results of Growth Optimization .............................. 49
5.3.1 Heater Temperature ................................. 49
5.3.2 Target-to-Substrate Distance ............................ 50
5.3.3 MPDR-610 Oxygen Source ............................ 51
Deposition of YBCO Films in Molecular Oxygen .................. 53
Deposition of YBCO Films Using 3 Microwave Cavity .............. 63
Deposition of YBCO Films Using an MPDR ..................... 66

Chapter 6

Summary and Conclusions

Conclusions and Suggestions ................................ 78
LIST OF REFERENCES ................................... 80

vi

Table 3.1:

Table 3.2:

Table 5.1:

Table 5.2:

LIST OF TABLES

Lattice mismatch for c-axis oriented YBCO on various
substrates (after Norton, reference 34). .................... 27

Reflection and absorption coefficients for various lasers on
YBCO (afier Wu, reference 49). ......................... 3O

YBCO films with Tc(R=0) between 73 and 90 K. Films were
deposited using molecular oxygen. ....................... 54

Comparison of films deposited in molecular oxygen and films
deposited in activated oxygen from a tunable-stub microwave

cavity. .......................................... 68

vii

Figure 1.1:

Figure 1.2:

Figure 1.3:

Figure 1.4:

Figure 1.5:

Figure 1.6:

Figure 1.7:

Figure 2.1:

Figure 2.2:

Figure 4.1:

Figure 4.2:

LIST OF FIGURES

Meissner effect in bulk superconducting material.
(after Kittel, Reference 6) ............................. 5

Magnetization vs. applied magnetic field for (a) Type I and
(b) Type II superconductors. (afier Kittel, Reference 6) ......... 6

(a) An SIS Josephson junction tunnelling device and (b) the
corresponding I-V characteristics at T=0. (after Decroux,
Reference 14) ..................................... 7

Flux Quantization Effect (after Decroux, Reference 14) . . . ...... 9

Time-line showing the increase in critical temperature, TC(R=O).
(after Robins, Reference 15) ........................... 11

Diagram of (a) the perovskite unit cell (b) the overall structure of
YBCO and (c) the Cu-O conducting planes sandwiching a single Y-

plane. (after Sleight, Reference 1) ........................ 12
Energy band diagram above and below the superconducting

transition. (after Decroux, Reference 14) ................... 15
Schematic diagram of a laser ablation deposition system. ........ 18

Schematic diagram of the laser/plasma/target interaction:

(A) unaffected target(B) evaporated target matter.(C)dense

plasma absorbing laser radiation (D)expanding plasma transparent

to the laser beam.(after Singh, Reference 23) ................ 24

Schematic of the laser ablation system incorporating the MPDR-610
oxygen source. ..................................... 36

(a) Photograph of the final laser ablation chamber design,
incorporating the MPDR and (b) a schematic of the MPDR-610

ion/free radical source. ............................... 37

viii

 

Figure 4.3

Figure 4.4

Figure 5.1:

Figure 5.2:

Figure 5.3:

Figure 5.4:
Figure 5.5:
Figure 5.6:

Figure 5.7:

Figure 5.8:

Figure 5.9:

Figure 5.10:
Figure 5.11:

Figure 5.12:

Figure 5.13:

The influence of substrate temperature on TC(R=O) for SrTiO,, Mgo,
and Si + ZrOz. (After Blank, Reference 66) ................. 41

Oxygen partial pressure vs. temperature plot showing critical
stability limits for three phases of YBCO.
(after Hammond, Reference 78) ......................... 44

Ratio of maximum forward power to reflected power of the

MPDR-610 vs. oxygen partial pressure. The substrate heater is

placed at various distances from the MPDR. The input power

is 200 Watts of 2.45 GHz microwave radiation ............... 52

Resistance vs. Temperature for YBCO on (100) YSZ.
The substrate temperature was 780 °C. .................... 55

Diagram of the patterned YBCO microbridge. The I, V
designations represent the 4-point probe contacts .............. 56

Critical Current 0,) vs. Temperature for YBCO film sample #CLA-Z 58
Profilometer measurement of a 40 um wide YBCO bridge. ...... 59
Plume Geometry in the laser ablation deposition system. ........ 60

Relative stoichiometry vs. position for YBCO film deposited in
molecular oxygen. .................................. 61

X-ray diffraction pattern of YBCO film on (100) YSZ. ......... 62

Scanning electron micrograph (SEM) images of the surface of
a YBCO film on YSZ. ............................... 64

Resistance vs. Temperature of YBCO film on (100) YSZ
deposited using a microwave cavity oxygen source. ........... 65

Critical Current (1,) vs. Temperature for YBCO sample
#LLA-73. The IC value at 77 K is 1.1 x 10‘5 A/cmz. ........... 67

Normalized resistance vs. T (K) for two samples prepared in
(a) 02 at 75 mTorr, Tc(0)=76K and (b) 75 mTorr activated
oxygen, Tc(0)==8lK. ................................. 69

Resistance vs. temperature for YBCO on YSZ. The film was

grown using the MPDR-610 oxygen source at 180 watts, 4mTorr,
and Twm = 815°C. ................................. 70

ix

 

Figure 5.14:

Figure 5.15

Figure 5.16

Figure 5.17:

Critical current density (I,) vs. temperature for sample
#CLA—131 prepared using the MPDR-610. The IC value at 77 K
is 2.9 x 105 A/cm2 .................................. 72

X-ray diffraction pattern showing c-axis orientation of YBCO
on (100) YSZ. The sample number is CLA-l31, prepared using
the MPDR-610 at 4 mTorr. ............................ 73

Scanning electron micrographs of film #CLA-131 at (a) 2 pm
and (b) 1 um resolutions. .............................. 74

Resistance vs. Temperature curve for YBCO on YSZ deposited
at 10 mTorr 02. The film exhibits semiconductive behavior. ..... 75

Introduction

Since the discovery of the ceramic high-temperature superconductors (HTSC) in
1986, scientists and researchers have been striving to integrate HTSC materials with
microelectronics. Before HTSC's, the study of superconducting materials took place at
liquid helium temperatures (4.2 K) which is an expensive refrigerant. The discovery of
YBaQCu307, BiSrCaCuO, and other HTSC ceramics have opened the door for the
development of inexpensive, nitrogen-based refrigeration and possibly on chip
refrigerators. Possible microelectronic applications exist in the form of sensitive magnetic
field detectors or gradiometers, infrared detectors or bolometers, high-speed devices such
as Josephson junctions, active and passive microwave devices, and superconducting
interconnects for VLSI applications. Before these applications can be realized, several
superconductor/semiconductor thin-film fabrication issues must be addressed. These
issues include finding compatible substrates, etchants, photoresist materials, and
passivation materials. Also important to HTSC microelectronic integration is improving
overall film quality through the development and optimization of the thin film deposition
techniques. Present YBCO deposition techniques include laser ablation, electron beam
evaporation, sputtering, and metal-organic chemical vapor deposition (MOCVD). Each
deposition process has its own advantages and disadvantages which make it suitable for
the particular HTSC thin film application. The laser ablation technique was used for our
experiments and its advantages, disadvantages, theory, and details of the deposition

conditions will be discussed.

2

The most promising HTSC ceramic for microelectronic application is YBaZCu3O7
or "123". It is a metastable compound which is often referred to as a solid solution since
it can exist over a broad range of compositions [1]; therefore, we use the term "phase"
instead of "compound," as in "123-phase." Most of the copper oxide superconductors are
metastable, but 123 has emerged as the dominant material in thin-film research for a
couple of reasons: (1) YBCO has been the easiest material to obtain single-phase films
and (2) it has the highest measured critical current densities, J, [2]. Using current thin-
film deposition techniques, most other HTSC ceramics form more than one stable phase.
Some of these phases may be non-superconducting or superconducting at lower
temperatures. This leads to poor overall film qualities such as reduced critical
temperature, T,, or semiconducting behavior. Conversely, single-phase YBCO films have
been quite easy to fabricate. High I, , which is important for device applications has been
measured to be more than 5 x 10‘ Mom2 at 77 K for some YBCO thin films[3]. Other
HTSC ceramics have J, values 1-3 orders of magnitude 1ess[4,5].

Conventional laser ablation processing conditions are 150-200 mTorr O2 partial
pressure and the substrate temperature >650°C. Understanding the effects of lowered
(<650°C) substrate temperatures and lowered oxygen partial pressures (<40 mTorr) is
important for improving YBCO film quality. Under the conditions of lowered substrate
temperatures and lowered oxygen partial pressures, molecular oxygen is insufficient for
the necessary formation of the Orthorhombic-II superconducting phase (X=6.93) of
YBaZCu3Ox. This phase has recently been grown at lowered pressures by laser ablation
with atomic oxygen. Utilizing electron cyclotron resonance (ECR) magnets, a microwave

plasma disk reactor (MPDR) can produce high concentrations of atomic oxygen

3

(>10“/cm3) at low pressures (<10 mTorr). The goals of this research were directed
toward designing a laser ablation system, optimizing YBCO laser-ablation growth
conditions at various molecular oxygen pressures and optimizing low-pressure (<10
mTorr) growth using atomic oxygen from an MPDR source. We successfully deposited
YBCO films using the MPDR at low oxygen partial pressures (4 mTorr, 750 °C) and
films exhibited zero resistance at temperatures as high as 79 K and the measured critical

current was as high as 2.9 x 105 A/cm2 at 77 K.

Chapter 1

Basics of Superconductivity

1.1 Low-Temperature Superconductors and Superconducting Phenomena
1 .1. 1 Zero-Resistance State

The phenomenon of a material exhibiting zero electrical resistivity at a finite but
sufficiently low temperature has intrigued scientists since the observation of
superconductivity in mercury at 4.15 K by Kamerlingh Onnes in 1911. When a material
achieves the superconducting state, the transfonnation is said to occur at the critical
temperature, T,, At or below T,, a loop of superconducting material can carry the same
electrical current for extremely long lengths of time[6]. Essentially, the material becomes
a 'perfect' conductor. The decay of supercurrents is so slow, that File et a1. predicted they
would last up to 100,000 years in a solenoid of Nb_,,Zr_25 [7]. The electrical properties
are also characterized by a critical current, 1,. This value is the amount of current per unit
area which can be passed through the superconductor before it goes into its normal state.
1.1.2 Meissner Effect

Aside from their electrical properties, superconducting materials also have unique
magnetic properties. Cooling a superconductor below its T,, any magnetic flux present
before the transition is expelled from sample. This characterization of the
superconducting state is called the Meissner effect and is illustrated in figure 1.1. As can
be seen, the bulk superconducting material has zero magnetic induction in its interior,

which is equivalent to a magnetic susceptibility, x, equal to -1 in the complete

4

superconducting state.
In the presence of higher magnetic fields, the diamagnetic property (or Meissner

effect) breaks down at a certain critical magnetic field, H,. Type-I and Type-II

// s.-.
W/ \V»

liwiia

\ a
\\\ W/

T<Tc

Figure 1.1: Meissner effect in bulk superconducting material.
(after Kittel, reference 6)

superconductors are classified based upon the criteria of the magnetization curves. Type-I
superconductors or 'sofi' superconductors exhibit a vertical drop in magnetization at the
critical magnetic field, H, (Figure 1.2a)[6]. Lead and Mercury are examples of Type-I
superconductors. The critical field approximately follows the temperature dependent
relationship:

H,=H,[1-(T/T,)2]

Type-II superconductors or 'hard' superconductors retain their zero-resistivity state

6

in high magnetic fields. This is due to flux pinning, whereby normal and superconducting
regions exist in a 'mixed‘ state or vortex state. Flux lines are prevented from moving by

imperfections such as precipitates, grain boundaries, and dislocations. As illustrated by

TYPO " /

. 411M

 

 

 

Figure 1.2: Magnetization vs. applied magnetic field for (a) Type I and (b) Type II
superconductors.(afier Kittel, Reference 6)

figure 1.2 (b), the onset into the vortex state occurs at the lower critical field, H,” and the
transition to normal state occurs at the upper critical field, Ha. The new high-temperature
superconducting ceramic (HTSC) materials are type-II superconductors. The vortex state
makes type-II superconductors more exploitable for applications such as superconducting

bolometers.

Two parameters useful in characterizing superconductors are the London

penetration depth, 3., and the coherence length, cg. These parameters are very important

7

for exploitation in Josephson junction devices. The penetration depth is a measure of the
depth of penetration at which an applied magnetic field decrease by a factor of e". A
typical experimental value is 500 A[6]. The coherence length is a measure of the distance
through which the superconducting wave function of supercurrents are coherent. Type-I
superconductors usually have a coherence length much greater than the penetration depth,

whereas a type-II superconductor generally has a larger penetration depth compared to the

coherence length.

1.1.3 Josephson Effect
Superconducting materials with sufficiently long coherence lengths are useful for

Josephson tunneling superconductor/insulater/superconductor heterostructure devices. A

thin insulating layer (10-20A) is sandwiched between two superconducting layers as shown

 

 

 

 

 

 

 

MlWMfHN“
d ' , ’
Je Ar " ‘7’
,2 Walled—p , ’
I“ /
I: —e> 4
g I 24/e
o /
(I) , ’
VOLTAGE
(b)

Figure 1.3: (a) An SIS Josephson junction tunnelling device and (b) the
corresponding I-V characteristics at T=0. (after Decroux, Reference 14)

 

8

in figure 1.3(a). When the device is cooled below T, and no voltage is applied across the
junction, Cooper pairs can tunnel across the insulating barrier, resulting in a junction
current with no applied voltage. This effect is possible since Cooper pairs (the main
mechanism for superconductivity in BCS theory) can travel a distance C without breaking
the pair [8]. The tunneling probability for Cooper pairs is given by
P(x) ~ e‘zx"
where x is the tunneling distance and d is the thickness of the insulating barrier [9,10,] l].
Longer coherence lengths are important for high Cooper pair tunneling. However, the
Cooper pair tunneling current is limited to the critical current of the junction, J,,. The
junction critical current, J,,, is a smaller quantity than the critical current of the

superconducting material, 1,. A generalized expression for this current is given as

= nA (T) A (T)
JCO 2e19,, can“ 21cc)

where A is the bandgap of the superconducting material and RI, is the normal state

resistance of the junction[ 12].

If we consider figure 1.3 (a), where the superconducting Cooper—pair wave

functions of the materials on either side of the barrier, ‘i’, and ‘P2, are given by

v1=ltr1 19“”

v, = is, is“:
where q», and (1’2 are the Josephson phase differences and AP, F and H’ZF are the densities

of the C00per pairs on each electrode[13]. These densities can be assumed to be constant

for not too large currents. The junction current can now be represented by J,=J,sin(<l)1 -

 

9

(1),). As d—+0 and l‘I’ l2 ->0, the junction current is given by

J,=i’£§ ivP (aw-23A)
m ' C
where m' is the effective mass of the electron, h is Plank's constant, and A is the area of
the junction.
If the magnitude of the applied voltage, N l, is greater than ZA/e (where 2A is the

minimum energy to break a pair), quasi-particle tunneling dominates. This effect is

shown by the hysteric I—V characteristics of figure 1.3(b). The junction now behaves

similar to a normal-state tunnel-junction. These I-V characteristics make Josephson

junctions usefirl for high-speed switching devices.

 

 

B-noo

Figure 1.4: Flux Quantization Effect (after Decroux, reference 14)

 

10

l. 1 .4 F lux-Quantization Effect

The effect of flux quantization is useful in superconducting loop structures such
as SQUIDs (Superconducting Quantum Interference Devices). By the flux-quantization
effect, the magnetic flux, B, within a loop of superconducting material is forced to be a
multiple of the flux quantum, 6D,. All the Cooper pairs in the ring of superconductor have
the same superconducting wavefunction and are related to another point on the loop by
phase. The flux-quantization effect is illustrated in figure 1.4, where the magnetic flux

lines are shown to be trapped in the hole of the superconducting loop.

Magnetometers made from SQUIDs have sensitivities several orders of magnitude

higher than other commercial magnetic sensing devices [14].

1.2 High-Temperature Superconductors
Most of the research in the years after Onnes' discovery (1911 - 1964) was on the

Niobium based alloys. The highest T, was 23 K, still well below the limit of

conventional cryogenic technology, 77 K. Cryotechnology above 77 K is relatively

inexpensive since it can be based on nitrogen gas. Oxide superconductors were

discovered in 1964, but the major breakthrough did not come until 1986 with the
discovery of the (La2_,,Ba,,)CuO4 or 2-1-4 ceramic superconductors which exhibited
superconducting behavior at 35 K [1]. Then in 1987, Paul Chu of the University of
Houston discovered the milestone superconductor, YBCO, with a T, of 94 K. This was
extremely significant since it is the first material to break the 77 K barrier [15,16]. Other

oxide superconductors were then developed with T,'s as high as 125 K as seen in the

timeline of figure 1.5.
YBaQCu307 (abbreviated: 1-2-3) is presently the most widely studied high-

 

 

ll

 

 

 

 

 

 

 

 

 

160
. 140 -
g TlBoCoCuO
32, 120 -
D BiSrCaCuO
E 100 _ YBCO
LLJ
% 80 I —77 K, Liquid Nitrogen ——
P:
2" 60 _ LaSrCuO
g
E 40 -«
Q N

20 _ NbN Nbssri Nb-Al-Ge $991.1 ”300W

Hg Pb Nb ,__1—““
.4 -
o r , I r
1900 1920 1940 1960 1980 2000

TIME (Years)

Figure 1.5: Time-line showing the increase in critical temperature, T,(R=0).(after
Robins, Reference 15)

temperature superconductor (HTSC) for device applications. This is true because thin
films of YBCO are easy to fabricate than other HTSC materials and have properties more
promising for microelectronic device applications. The complex cuprate superconductors
shown in figure 1.5 exhibit superconductivity at temperatures up to 125 K, but are much
more difficult to fabricate because of the existence of multiple stable phases.

1.2.1 Physical Properties
YBCO films have the highest critical currents, J,, to date up to 5 x 10‘ A/cm2 at

77 K[17]. Most of the research is directed toward YBCO films instead of bulk material

because reported J, values for bulk YBCO are only 40 - 3000 A/cm2 at 77 K[S]. Upper

12
critical field values of YBCO films at 77 K range from 20 to 40 Tesla, which make them

good for practical device applications. 123 has a coherence length of approximately 7 A
perpendicular and 34 A parallel to the c-axis direction[16]. Because of the short
coherence length as compared to low-temperature superconductors, grain boundaries and
other lattice defects can act as weak links (Josephson junction with a larger barrier),
reducing J, values. Effort is underway to fabricate a-axis oriented films for better
Josephson junctions since the coherence length is longer parallel to the c-axis.
1.2.2 Crystal Structure

The YBCO structure can be described as a triple unit perovskite cell with yttrium
and barium ions in alternating layers and copper atoms on the intervening planes. A
diagram of the general perovskite unit cell is shown in figure 1.6(a) and the YBazCu3O7
structure is shown in figure 1.6(b). This overall structure shown in figure 1.6 (b) is the
orthorhombic-II superconducting phase (T,(R==0)==94 K) with lattice parameters of a=3.82

A, b==3.89 A, and c=l 1.68 A[13]. 123 also exists in a tetragonal, non-superconducting

 

 

 

 

 

 

 

 

 

 

Figure 1.6: Diagram of (a) the perovskite unit cell (b) the overall structure of YBCO
(c)the Cu-O conducting planes sandwiching a single Y-plane.(after Sleight,reference 1)

13

phase and an 'oxygen-deficient' orthorhombic-l phase which has a lowered T, of
approximately 50 - 60 K[17]. The tetragonal structure is also referred to as 'oxygen
deficient' and behaves as a semiconductor.
Fabrication of films or bulk YBaZCu3O7 usually involves going through two phase
transformations. The tetragonal phase, YBazCu3O7,X (X = 6.0 - 6.2), is the most stable
phase. During cool down, the 123 sample is cooled slowly in oxygen to bring the oxygen
content up to 7. The cooling times vary from one hour to several hours, and the resulting
YBCO orthorhombic-II phase has a T, above 90 K. If the highly-stable tetragonal phase
is formed and then cooled rapidly in oxygen, insufficient oxygen incorporation may result
in formation of the orthorhombic-1 phase (X=6.2 to 6.5) which may have broad zero-
resistance transitions, reduced T,'s, or non-superconducting properties.
1.3 Theories of Superconductivity
1.3.1 Electrical Properties
The structure of YBa2Cu3O7 shown in figure 1.6(b) illustrates the two Cu—O planes
which sandwich the yttrium plane. Conduction is believed to occur due to the motion of
holes in the Cu-O sheets. Although there is some superconducting wave firnction (w)
overlap in the c-axis direction, most conduction takes place in the a-b plane. It is
believed that conduction occurs in the a-b plane for two reasons: (1) the oxygen atoms
in the c-plane are too far apart or completely absent in the c-direction (Y-layer) and (2)
the oxygen atoms in the c-axis direction are further away than the oxygen atoms in the
a-b axis layers. This simplifies to a quasi-two-dimensional conduction model.
The density of carriers (holes are believed to be the dominant caniers) in YBCO

is approximately 102‘lcm3 which is low compared to conventional superconductors [l6].

l4

Substitution of yttrium with other elements in the YBCO lattice changes the
superconducting properties by changing the lattice spacing between the Cu-O sheets. The
properties of the material can change from varying superconducting properties to
semiconducting properties. This substitution method is analogous to modulation doping
in semiconductor superlattices.

1.3.2 BCS Theory for Low-Temperature Superconductors

Early theories of superconductivity were posed by Gorter and Casimir (1934),

London (1935), Ginsburg and Landau (1950), Frohlich (1950), and Pippard (1953). The
major theory was posed in 1957 by Bardeen, Cooper, and Schrieffer which is known as
the BCS theory of superconductivity. Previous models were unable to integrate the
presence of an energy gap and the similar ground state of all superconducting electrons,
the electron lattice interaction, large scale coherence, and the 'boson' theory of electron
interaction.

The electron-lattice interaction posed by BCS theory takes place as follows
[6,8,13]: At T < T,, an electron with 8 z8, moves through the lattice and comes within
10" cm of an ion. The electron cloud of the ion is repelled by the electron and the ion
is slightly polarized. The ion is then weakly attracted to the electron and the ion deforms
the lattice. This slight deformation of the lattice is equivalent to the absorption of a
phonon. These phonons are then emitted by the ions.

We now let 1., and Mv) be the wave vectors of an electron interacting with the ion

and of the emitted phonon respectively. A second electron with wave vector 11,, (with
spin and momentum opposite of the first electron) comes along and repels other electrons

while emitting a phonon with wave vector — 1(v). The magnitude of the phonon wave

 

15

vectors are equal but are in opposite directions. The emissions of phonons reduces the
nominal energies of each electron to a value of 8 - 8f - 8J2, where *3, is the bonding
energy of the Cooper pair. Essentially, the second electron adjusts itself to a lower
energy to take advantage of the lattice deformation. The net result is a lower repulsion
between the two electrons and the formation of paired states or Cooper pairs. This effect
is illustrated in the energy band diagram of figure 1.7. In the figure, the value 2A is the
energy gap and A = 8,. The electrons which interact with the filled electron shells of the
ions are called quasi-particles and are shown in the conduction and valence bands of the

energy band diagram.
In this model, more than one electron can occupy the same energy state. Hence,

the Cooper pairs are known as 'bosons' since their energies follow Bose-Einstein statistics.

2A

2
i
i,

 

OO COOPER PAIRS

o . QUASIPARTICLES

Figure 1.7: Energy band diagram above and below the superconducting
transition.(afier Decroux, Reference 14)

16
All the electrons forming bosons have the same energy and momentum and can be
represented by a single wave function, ‘1’ = pmele, where p is the charge density and O is
the phase common to all electrons in the superconducting state.

Another accomplishment of the BCS theory of superconductor electron-lattice
interactions was that it helped to explain the isotope effect. It was found that H, and T,
0: 1/M “2, where M is the mass of the isotopes. The BCS theory was also used to predict
T, in low-temperature superconductors:

T, = 1.14(OD)exp[-1/(U o D(sf))]
where OD is the Debye temperature, U is the electron-lattice interaction, and D(ef) is the
density of states at the Fermi level.
1.3.2 RVB Theory and Goddard's Magnon Theory for HTSC

The BCS theory did not hold up when attempting to predict the behavior of
ceramic HTSC. It could not relate measured band-gap values to critical temperatures in
HTSC. The 'Resonant Valence Bond' theory proposed by Philip Anderson attempts to
explain ceramic superconductivity. He suggested that a resonating structure exists
involving a 'superexchange' of electrons between copper ions. The idea of hole carriers
was also proposed as channel by which the resonant exchange takes place.

Another theory for ceramic superconductivity is proposed by William Goddard.

It results in large scale ordering, based on 'Magnon-pairing'. The ordering is suggested

to be a result of the interaction of electron pairs with pairs of copper ions, causing an

alignment of their magnetic moments about the oxygen atoms.

Chapter 2
YBCO Thin-Film Deposition
2. 1 Introduction
A major portion of HTSC research has been directed toward their integration into
microelectronics. To realize this integration, reproducible methods for fabricating HTSC
thin films have to be developed. Furthermore, the film quality (crystal orientation,
smoothness, J,, T,, etc.) should suit the particular application. For example, c-axis
oriented films are desired for passive microwave striplines and filters where low-loss
conduction is desired in the plane parallel to the substrate[l8]. C-axis oriented films are
also useful for planar magnetometer applications because the magnetic penetration depth
is shorter in the a-b plane[l9]. The a-b axis oriented films have possible applications in
monolithic vertical Josephson junctions where longer coherence lengths are necessary.
Each thin-film deposition process can produce films which have physical properties which
are better or worse for the particular device application. Laser ablation (the process used
in this research) is one of the three main YBCO thin-film deposition techniques. The
other two main techniques are sputtering and electron beam co—evaporation. Descriptions,
advantages and disadvantages of each deposition technique will be discussed, but most
emphasis will be given to laser ablation.
2.2 Laser Ablation, Sputtering and Electron-Beam Evaporation
Laser ablation has been established as a versatile, single-source, deposition
technique, allowing one-step fabrication of thin-films and multilayered structures without

breaking vacuum. This technique is not only promising for YBCO but also for thin-film

l7

18

processing of other materials. Semiconductors, insulators and HTSC materials have been
deposited onto various substrates using laser ablation. These materials include SiC, Si,
Zr02, MgO, SrTiO3, YBCO and other HTSC materials.

Laser ablation involves firing nano-second pulses from an excimer laser onto a
stoichiometric target. A "plume" or plasma forms and deposits material normally onto
the facing, heated substrate (see figure 2.1). Stoichiometry is preserved because the
deposition of the target elements is not determined by vapor pressure of the emitted
species, but instead by the speed with which the elements leave the target. In the case
of YBCO deposition, film growth occurs atomic-layer by atomic-layer from the arriving
metal oxides (YO, CuO, BaO)[2]. With the indiffusion of oxygen during YBCO cooling,
the laser ablation technique can be an in situ process in which the original target

stoichiometry is preserved.

In addition to in situ processing, another advantage to laser ablation is the high

02

 

 

 

 

 

 

 

 

 

 

Laser
:19 [I :3!)
\ ; a
Rotating

Substrate YBaCuO target
and heater

Vacuum Plume

port

 

 

 

Figure 2.1: Schematic diagram of a laser ablation deposition system.

l9

deposition rate as compared to other thin film processes. The laser ablation deposition
rate for YBCO has been recorded to be as high as 145 A/sec[20]. This rate is orders of
magnitude higher than sputtering (1-10 A/sec) and electron beam co-evaporation (1-5
A/sec). Disadvantages to this technique include the presence of micron-size surface
particles and the high cost and inefficiency of a laser as an evaporating source.

The single and multi-target sputtering techniques involve accelerating ions through
a potential drop. Molecules and species are liberated from the target surface using a
sputtering gun (100-150 V DC) or even an ECR beam[2l] in an argon/oxygen ambient.
Typical pressures are 400 mTorr argon and 200 mTorr oxygen. Advantages to this
method include reproducibility and J, values as high as 5 x 10‘ A/cm2[5]. The main
disadvantage is the low deposition rate (1-10 A/sec).

Electron beam co-evaporation is also a widely used technique for deposition of
YBCO thin films. This method uses three metal vapor sources which are evaporated by
electron guns. Stoichiometry and evaporation rates are carefully controlled using crystal
monitors or other spectroscopic techniques. An oxygen ambient of approximately 10'3 to
10" Torr is introduced via a tube near the substrate. One advantage to this method is the
ability to control stoichiometry. Disadvantages include compositional and rate changes
in the metal vapor sources with exposure to high oxygen pressures (>1mTorr), reduced
lifetime of the e-beam filaments with exposure to oxygen, low deposition rate (1-5 A/sec),
and line-of-sight deposition leading to film non-unifonnity over larger areas[22].
2.3 Theoretical Model of Laser Ablation

Characterizing the laser ablation process has been a complicated undertaking with

researchers using theoretical models and computer simulations. By studying the physical

20

deposition parameters of the laser ablation process, Singh and Narayan have developed
a simplified physical model [23]. Their model simulates the laser/plasma/solid interaction
and is summarized in the following paragraphs.

Singh and Narayan break the ablation process down into three regimes: (i)
interaction of the laser beam with the bulk target resulting in evaporation of the surface
layers, (ii) interaction of the evaporated material with the incident laser beam, and (iii)
anisotropic adiabatic expansion of the plasma. The last regime leads to the nature of the
deposition process. In the first regime, the high-powered, nanosecond laser pulse (up to
800 mJ/pulse, .5 - 10 J/cmz, 8 - 20 ns pulse duration) is fired at the target. The result is
the melting or evaporation of the target surface layers. Using heat balance, the amount
of evaporated material per pulse is calculated as

Ax, = (1 - R)(E - E,,,)/(AH + CAT) (eqn. 1)
where Ax, R, AH, C,, and AT are the evaporated thickness, the reflection coefficient of the
laser on YBCO, volume latent heat, volume heat capacity, and the maximum temperature
rise respectively. E,,, represents the energy threshold or the minimum energy where
appreciable evaporation is observed. For excimer laser irradiation, E,,, z .05 - .4 J/crn2 for
YBCO targets and 3.5 - 4.0 J/cm2 for silicon. Equation (1) is valid for conditions where
the thermal diffusion distance (2Dr)"2 is larger than the absorption length of the laser
beam in the target material, 1/a,. In this equation, D is the thermal diffusivity, and t is
the pulse duration.

The second regime is the interaction of the evaporated material with the incident
laser beam resulting in an isothermal plasma formation and expansion. They derive the

initial expansion of the three orthogonal plasma edges formed during this regime. The

21

material evaporated from the target is heated further by the absorption of laser radiation.
Neutral species, electrons and some positive ions are present in the initial evaporated
ionized-vapor. The primary absorption mechanism for a plasma is the electron-ion
collisions. Singh and Narayan represent the absorption by an inverse bremsstrahlung
process, involving a free electron absorbing a photon. The absorption coefficient, (1,, of
the plasma is expressed as
a, = 3.69 x 10’ (Z3n,2 / TV’v’) [1 - exp(-hv/kT)] (eqn. 2)

where Z, n,, and T, are the average charge, ion density, and temperature of the plasma,
respectively. The terms h, k, and v are the Planck constant, the Boltzman constant, and
the frequency of the laser light. This equation assumes that the plasma frequency is
smaller than the frequency of the laser light. For an excimer laser with a wavelength of
308 nm, the laser frequency is 9.74 x 10”sec“. For the same plasma frequency, (up, the
corresponding electron density is 1.2 x lOzz/cm3. Since this value of electron density is
high, and it can be assumed that reflection losses by the plasma are insignificant for
excirner-laser-generated plasmas. The [1 - exp(-hv/kT)] term represents the losses due to
stimulated emission. The absolute value of the plasma absorption coefficient is quite
difficult to compute since it depends on so many parameters.

The plasma absorption decreases further away from the target, at the leading edge
of plasma. The high expansion velocities (10S - 106 cm/sec) decrease the ion and electron
densities, making it transparent to the laser beam in this region. Figure 2.2 diagrams the
laser/plasma/target interaction and shows a thin region near the target which constantly
absorbs laser radiation. The plasma which absorbs the laser radiation is simulated as a

high-temperature, high-pressure gas, initially confined to small dimensions and then

22

allowed to expand in vacuum. The authors then give the density of the plasma at any

point (n(x,y,z)) and at time t as

NTt x2 I y2 22
.3
21/219”er Y(t)ztc) €pr Meg” )

. . .t = " " '
HIXYZ ) 2X(t:)2 21/(c)2 22m2

 

 

for t.<_r, where NT is the total number of evaporated particles at the end of the laser pulse
(t=r). X(t), Y(t), and Z(t) represent the orthogonal directions of the expanding plasma and
correspond to the distance at which the plasma density decreases to 60.65% of the
maximum density. The authors assume that the plasma behaves as an ideal gas; thus, the

pressure is related to the density by the equation (P = nkTo) and can be expressed as

NTckT,
21/2n3/2rxt c) r( c) zt 1:)

X2 .- yz _ 22 .4
2X(t)2 arm2 22(c12Heqn )

 

P(x,y,z, t) = expl-

 

for 151:, where T0 is the isothermal temperature of the plasma. Based on a previous
argument for gaussian density profiles[24], the velocity of the plume species should be

proportional to the distance from the target. This can be represented by

x (”“31+ 1 (”(9) z dz(t)k(eqn.5)

W"""""'c)=.r(t) dc Y(t) dc 1+2”) dc

where dX/dt, dY/dt, and dZ/dt refer to the expansion velocities of the plasma edges X,
Y, 2, respectively. The continuity equation governs the expansion of the plasma and can

be expressed as

a mNTt

--§Ejvpdv=isp (“r-m dA-a—t t (eqn. 6)

 

where V denotes the volume, and the surface enclosing V is denoted by S. The
differential area element is denoted by dA, and N is the unit normal vector. In equation

6, p corresponds to the density of the fluid and m to the mass of the atomic species. The

23

last term of equation 6 shows the injection of atomic species into the plasma. The

equation of motion can be expressed as:

6‘7 —_§£ "‘. “ "‘ “‘. :— '_ =
[V[p-6E+vdt+p(vV)v+v(vVP)+p(Vv)v+VPldV O(eqn.7)

Substituting the equations 3, 4, and 5 into equations 6 and 7, the authors anive at a

solution given by

_1___d__Z d2=1<zl

7:37,} dc? ;°<eqno 8)

             

 

X(c) 1%%+%1 =Y(c) 1% g—{fi
for £31. This equation determines the initial expansion of three orthogonal plasma edges.
The initial dimensions of the plasma are of the order of mm in the transverse direction,
whereas in the perpendicular direction they are less than him.

In the third regime, the adiabatic plasma expansion, the solution which controls
the expansion can be found by substituting the velocity, density, and pressure equations

into the differential equations (5 and 6), the adiabatic equation of state, and the equation

of temperature. This equation is given by:

73:] k_r_'_,[ X,Y,z, 11-:

M x(c)r(c)z(t)

(1:1’

 

               

X(CH-g— :5]: Y(C)

 

for t>r, where y is the ratio of specific heat capacities at constant pressure and volume.
The terms X,, Y,, and Z, are the initial orthogonal edges of the plasma at the end of the
laser pulse (t—=t). This equation is in good agreement with the actual, elliptical shape of
the expanded plume. Our experiments indicate that the position of the expanded-plasma-

edges relative to the substrate is directly related to film stoichiometry.

24

 

ll ll ll

 

   

 

 

 

 

 

BULK TARGET PLASMA CLOUD

Figure 2.2: Schematic diagram of the laser/plasma/target interaction:(A) unaffected
target(B) evaporated target matter.(C)dense plasma absorbing laser radiation
(D)expanding plasma transparent to the laser beam.(afier Singh, reference 23)

2.4 Novel YBCO Thin Film deposition Techniques
2.4.1 Novel Laser Ablation Techniques

Several modifications to the laser ablation technique have been employed to
achieve successful YBCO growth at lowered temperatures (<650 °C). A high-voltage
accelerating ring has been used in conjunction with laser ablation to sustain the laser-
induced plasma near the substrate[25]. This method has yielded films with T,(R=0) as
high as 85 K at a substrate temperature of 400 °C. Substrate biasing has also been
employed to lower the substrate temperature. Atomic species are attracted to the biased
substrate and increase the amount of oxides necessary for YBCO growth. Films showed
improvements in T, (with 1300, i500 V applied to the substrate) over the temperature

range of 630 to 67‘ °C. Another technique uses a pulsed oxygen source to deposit films

25

in a low background pressure with (10“ - 10'3 Torr) a high oxygen pressure near the
substrate[27].
2.4.2 Activated Oxygen Enriched Film Deposition

Kwo et a1. [28] have reported remarkable increases in T,(0) using molecular beam
epitaxy, a differential pumping scheme and activated oxygen produced by a microwave
cavity (2.45 GHz excitation frequency). The oxygen partial pressure near the substrate
was in the mid 10" Torr range. Activated oxygen enhanced films on (100) MgO
exhibited R=0 transitions at temperatures as high as 89 K.

Greer conducted an investigation on reduced-oxygen partial pressure laser-ablation
depositions [29]. He used laser ablation with a microwave cavity (non-ECR, 500 watts
input power) to deposit YBCO films on (100) SrTiO3 and (100) sapphire at .5 mTorr
using a quartz flow tube. Measurement of the atomic oxygen flux near the substrate was
5 x 10“ atoms/cmZ-sec at .5 mTorr. T, values were as high as 65 K on SrTiO3, and Greer
notes that attempts to increase the atomic oxygen flux at the substrate were deleterious
to the electrical properties of the films. Post annealing also failed to affect film quality.
This result is supported by work on e-beam co-evaporation deposition of YBCO with
ozone/oxygen mixtures. It was found that beyond a 2% mixture of ozone in molecular
oxygen, no improvement in electrical properties was found [22].

Yamamoto et al. [30] have used the Wavemat MPDR-610 for post-deposition
oxidation of 90° off-axis sputtered YBCO films. Their results showed a decrease in
resistance by more than one order of magnitude (4000 ohms to 200 ohms at room
temperature) in the presence of the ECR plasma, and no information on T,(R=0) values

are given. Using-gtomic oxygen in conjunction with e-beam evaporation, Hurnphries et

26

a1. [31] have observed the formation of the orthorhombic phase of YBCO below the
pressure-stability limit of YBCO in 02-

Recently, O'Keefe et al. have devised their own ECR ion/free radical source and
have used it in conjunction with the laser ablation process [32]. Measurements indicate
ion densities are above 10‘°/cm3 [33] when the magnets were in place. Films produced
using the ion source without the ECR magnets in place (Oz+ + 0' present) produced films
exhibiting zero resistance as high as 56 K. The deposition conditions were: 193 nm laser,
substrate and ECR source 5 cm from target, 10 mTorr oxygen partial pressure and a
substrate temperature of 640°C. With the ECR magnets in place (high 0' concentration),
T,(R=0) values are as high as 69 K under the same deposition conditions. At 690°C and
10 mTorr, T,(R=0) values are as high as 78 K for the ECR and non-ECR depositions,
which seems to indicate no improvement using an ECR plasma at this temperature and
pressure. One of our goals is also to use ECR oxygen plasma from an MPDR to deposit

YBCO films at low-pressures (<10 mTorr).

Chapter 3
YBCO Laser Ablation Deposition Parameters
3. l Substrates
Choice of an appropriate substrate material is important for the growth of quality,

c- or a-axis oriented YBCO films. Lattice matching, orientation, interfacial reactions,

Table 3.1: Lattice mismatch for c-axis oriented YBCO on various substrates (after
Norton, Reference 34).

 

 

 

 

 

 

Material Lattice parameter f=lattice mismatch
(nm)
MgO ' .421 9.7%
SrTiO3 .391 2.3%
LaAlO3 .378 -l.l%
YSZ[100] // YBCO[100] .516 <0.2%
YSZ[110] // YBCO[100] 5.9%

 

 

 

 

cost, and deposition technique must all be considered. The most commonly used
substrates for YBCO thin film deposition are MgO, LaAlO3, SrTiO3, A1203, yttria-
stabilized zirconia (Y 82) and several semiconductors with buffer layers of YSZ or SrTiO3.
Table 3.1 illustrates the lattice mismatch for c-axis oriented YBCO films on various
substrates[34]. In the third column, f = (a,-a,)/.5(a,+a,) where a, and a, are the lattice
constants of the substrate and of the YBCO overgrowth respectively.

Because the strain energy depends on the square of the lattice mismatch, epitaxial

growth occurs in orientations that minimize the lattice mismatch and in orientations of the

27

28

minimum energy configuration[35]. The large lattice mismatch of MgO leads to
misorientation of grains in the a-b plane. The YBCO lattice matches well with LaAlO3
and SrTiO3, but both are quite expensive and SrTiO3 has a poor dielectric loss tangent,
making it unsuitable for microwave applications [5]. Sapphire is more suitable for
microwave applications, but reacts with film and has a larger lattice-mismatch[36,37].
Substrate reactions are also a problem with the semiconductor substrates silicon and GaAs
[38,39]. For these substrates, researchers have turned to depositing buffer layers such as
YSZ, MgO or BaTiO3 [5,40]. In our earlier work, rapid thermal processing was also
employed to suppress the substrate reactions [41].
Deposition techniques such as in situ laser ablation and sputtering allow a variety
of compatible substrates, but when using other techniques such as MBE, reactions with
MgO and YSZ can arise. For example, barium zirconiate has been found to form and
consequently degrade films, while Mg substitutes for Cu in the YBCO lattice and reduces
T, [42]. We utilized the laser ablation technique, and YSZ was chosen as the
predominantly used substrate. YSZ has been proven to be a successful substrate in the
fabrication of high-J, (>5 x 10‘ A/cm2 at 77 K) and has a low dielectric loss tangent, 6,
at temperatures lower than 250 K (63.005)[43]. This makes it an excellent choice for
microwave device applications. Although the lattice parameter for the cubic YSZ (a=.516
nm) is much larger than that of YBCO (a=.382 run), the YBCO/substrate mismatch for
the (100) oxygen sublattice is <0.2%. Even in the YSZ[110] // YBCO[100] epitaxial
relationship (which is equivalent to the a & b axes of the YBCO film rotated 45° about
the c-axis) the mismatch is only 5.9%. With the laser ablation technique, temperature

ranges of 700 to 810 °C have been found to produce quality c-axis oriented films on

29

YSZ, and below 650 °C for a- and c-axis mixed orientations[44].
3.2 Target conditions

Laser ablation is a simple technique which requires only one source for deposition
of film material. A stoichiometric target of YBazCu3O7 is used in this case. The target
is usually prepared by the sintering process or can be purchased. Targets are usually .5"
or larger in diameter and a few mm thick. Target conditions can affect thin film
properties and some of which include: porosity of the target, smoothness of the target
surface, target rotation, and target-to—substrate distance. The first two, porosity and
smoothness, are qualities of the target and have an affect on the surface particle size and
density. Particulates are a common problem in laser ablated films with particle densities
of our films commonly on the order of 104 - 10°/cm2. Their presence has been shown not
to affect J, and T,, but they impede microelectronic device fabrication efforts.

The best quality YBCO films have been achieved when the target is rotated.
Scanning electron microscopy (SEM) images of the target surface after ablation (non-
rotated) show yttrium-rich cones, clusters, and other rough features[45]. These unwanted
clusters may be removed and deposited on the substrate during subsequent depositions.
In addition, the change in target surface morphology has been shown to reduce deposition
rates. If a low target porosity is used and the target is scraped with a razor blade or
sanded after each deposition, particle density can be reduced, but not eliminated.

The target-to-substrate distance also affects particle density, film thickness and
film uniformity [46, 47]. Target-to-substrate distance must be optimized with oxygen

partial pressure and laser fluence.

30
3.3 Laser Types and Conditions

There are various laser types used for ablation today: multiple-mode excimer lasers
(193, 248, 308, 351 nm), neodymium-yttrium aluminum garnet (Nd-YAG) lasers (1064,
533, 355 nm) and continuous wave (CW) CO2 lasers. Ablation efficiency and film quality
are important for evaluation of the laser. The excimer lasers have been the most
successful of the lasers for producing quality YBCO films. They can produce high energy
pulsed output at up to 800 mJ/pulse. In addition, the higher the absorption coefficient of
the laser on YBCO, the thinner the evaporated surface layer. The reflection coefficient
should also be small for efficiency of optical energy transfer. For example, using an Nd-
YAG laser (1.06 11m) at a fluence of .6 J/cmz, the melt front propagates to a maximum
depth of .55 um and the surface temperature does not reach the ablation temperature [48].
Excimer lasers have a much higher absorption coefficient and can reach ablation

temperatures at smaller melt depths. Table 3.2 summarizes the estimated reflection and

Table 3.2: Reflection and absorption coefficients for various lasers incident on
YBCO.(Afier Wu, Reference 49)

 

 

 

 

 

- W V '_ _r_r_ 0. .5 u 1

Carbon Dioxide ~ 10 0.5 0.75

Nd:YAG 1.064 1.2 0.18

Nd:YAG (M2) 0.533 1.5 0.14

Nd:YAG (M3) 0.355 1.7 0.12
Excimer

XeF 0.351 1.7 0.12

XeCl 0.308 1.9 0.12

KrF 0.248 2.3 0.13

ArF 0.193 2.4 0.15

 

 

 

 

 

 

31

absorption coefficients of YBCO for various lasers [49]. The absorption coefficient is
denoted by a. The table shows that the excimer lasers are the optimum choice in terms
of low-reflection of the YBCO target material.

Laser deposition parameters such as laser wavelength, repetition rate, and fluence
also affect film properties such as surface particle density, J,, and stoichiometry. It has
been found that the longer wavelength CW--CO2 and Nd-YAG lasers are less effective
than the shorter wavelength excimer lasers in producing quality laser ablated YBCO films.
Shorter wavelength lasers have been found to reduce particle densities [50,51]. A typical
laser ablated YBCO film deposited using a 193 nm excimer laser has less than 1% of the
surface area covered with particulates. An SEM of a YBCO film prepared under the
same conditions with a 1.064 um Nd-YAG laser shows 60% of the area covered with
surface particles. The results for the frequency doubled (534 nm) and frequency tripled
(355 um) Nd-YAG lasers are better with particle densities of 5% and 2% respectively.

Variance of repetition rate from 2 to 15 Hz showed particle density decreases with
decreasing repetition rate. A recent comparison of laser wavelength also shows the use
of shorter wavelength lasers improves critical current density[52]. Particle density is also
reduced with reduced laser fluence over the range of 3.5 to 1.7 J/cm2[47].

In addition to affecting particle density, laser fluence also has an impact on
stoichiometry and deposition rate. Based upon experimental and theoretical results, there
exists a threshold for appreciable evaporation of target material. Below this limit, there
is no evidence of surface melting. This energy density threshold, Em, has been estimated
using a quadrapole mass spectrometer [53]. E, for the excimer lasers was .279 J/cm2 for

351 nm, .141 J/cm2 308 nm, .050 J/cm2 for 248 nm, and .066 J/cm2 for 193 nm. For the

32
1.064 um Nd-YAG laser, the threshold energy density value was measured as .660 J/cmz.

Under low fluence conditions (<.5 J/cmz), there is only one velocity component of the
emitted species which has been measured to be $105 cm/sec[54].

Under increased irradiation (>1 J/cmz), the velocity distribution (measured by time-
of-flight (TOF) mass spectroscopy) shows a twin-peak distribution containing a larger
(>106 crn/sec, believed to be caused by scattering[44]) and a smaller velocity component
(210’ cm/sec). For typical YBCO deposition pressures (150-200 mTorr) the velocities are
unaffected by pressure initially, due to the formation of a shock wave in which the
species are almost collision free[55]. Afier the shock wave disappears, the
fragments/species approach the substrate and slow down due to the collisions with the
residual oxygen. The velocity of the emitted species can affect the amount of material
reaching the substrate in oxide form, a necessary condition for stoichiometric growth of
YBCO. Furthermore, at lower oxygen partial pressures (<10 mTorr) there is no shock
wave and the collisions due to residual oxygen are also reduced. The result is high
species/fragment velocities for a longer distances away from the target. These
experiments will be discussed in section 3.4 in greater detail since the velocity
components are also related to the oxygen partial-pressure. These results also suggest that
there may be a fluence limit, E, for optimal growth of YBCO films at a particular oxygen
pressure[56]. An upper limit or E,WAX has not been defined because lengthening the
target to substrate distance may compensate for increased velocities caused by high
irradiation conditions.

3.4 Oxygen partial pressure effects

Oxygen ambient pressure is critical for YBCO growth; it affects stoichiometry, T,,

33

J,, and film thickness. Many groups studying laser-ablation used spectroscopy to
determine the velocity of the plume species and content of the plume [57,58,59,60,61].
Recently, Sakeek et al. used TROSP (time-resolved optical absorption spectroscopy) to
study the expansion of the plume [57]. In 500 mTorr 02 the expansion velocities of the
molecular fragments/species (YO, Ba“, Ba, Y, Y‘, Cu) were found to be approximately
2.5x10’ cm/s at distances <l.5 cm from target and <2.5x104 cm/s beyond 1.5 cm.
Furthermore, a slower component was found to be ejected from the plume due to
evaporation of the target. Although these components initially show different velocities,
the fast component slows down in the presence of an oxygen atmosphere and forms a
front at the laser—induced plasma/oxygen interface. They concluded the species of
different velocities converge at the same oxidation front (which corresponds closely to the
tip of the luminous plume) and is the optimum substrate location for producing good
quality in situ YBCO films. The location of this luminous interface or plume 'edge' with
respect to the substrate is directly correlated with film stoichiometry and subsequently T,

and J,.

Chapter 4

Design and Construction of YBCO the Deposition System

4. 1 Introduction

The goals of this study were to design a laser ablation system, optimize the laser
ablation system, and investigate the effects of pressure on the quality of laser-ablated
YBCO films using the MPDR-610. The chamber was designed for versatility, for
accessibility, and with YBCO deposition considerations. The chamber was designed for
versatility because of the variety of modifications which can be used in conjunction with
laser ablation process. Accessibility is necessary since the chamber must be opened after
each deposition. Viewports and optical ports for spectroscopy and other in situ analysis
techniques were also considered in the chamber design. This chapter details our
modifications to the laser ablation system, including the incorporation of an MPDR
oxygen source. Design considerations for YBCO deposition parameters are also
presented.
4.2 Design Considerations
4.2.1 Chamber Design

The chamber body was made from a 12" long, 8" diameter stainless-steel, conflat
full-nipple. One end is blanked off with a 10" conflat flange; this is the bottom end of
the chamber. The top of the chamber is a 10" to 6" conflat reducing flange with a 6"
glass viewport. Handles were welded onto the top flange for easy alignment and removal.

A Viton gasket was used on the top flange since the top must be removed after each

34

deposition to access the film. Six conflat ports were welded onto the middle of the 8"
diameter tube at 45° intervals. A seventh port was welded onto the chamber, rotated by
60°. The chamber ports are used for the following purposes:

~pressure gage

-heater/thermocouple feedthrough

-laser beam entrance

-vacuum pump

-pyrometer

-target rotation

-MPDR oxygen source
A V4" stainless-steel tube with valve is also welded on the chamber for venting the system
with oxygen afier deposition. A top-view, schematic diagram of this chamber is shown
in figure 4.1. The distance between the target and the substrate can be varied from 3 to
6 cm. Including the ports, the total chamber volume is approximately 10.5 liters.
4.2.2 Microwave Plasma Disk Reactor

The MPDR-610 was incorporated into the laser ablation apparatus as depicted in
figure 4.2(a) and a schematic of the MPDR is shown in figure 4.2(b). The ECR plasma
disk region shown in figure 4.2(b) is 3 cm in diameter and emits a beam of ions and
atomic species. Electron cyclotron resonance (ECR) is a low-pressure coupling technique,
and the conditions for ECR to occur require the electron mean-free-path to be much
greater than the ECR orbit so electrons may then orbit many times between collisions

[62]. The ECR circular frequency orbit is given by the equation:

35

36

Lens 'I‘urbomolecular pump

 

 

 

Beam Infrared Pyromcwf

 

 

 

 

 

 

 

 

 

 

\ m
1* l 1:3)
L.

 

 

 

Substrate “ 1 '/ Rotating
and heater YBaCuO target
Plume ’MPDR Oxygen Source
and oxygen
plasma [________1 Vacuum Guage

 

 

 

Figure 4.1: Schematic of the laser ablation system incorporating the MPDR-610
oxygen source.

 

Figure 4.2: (a) Photograph of the final laser ablation chamber design, incorporating
the MPDR and (b) aschematic of the MPDR-610 ion/free radical source.

38

(1),, = eB/m,

where e = charge on an electron, B = strength of a static magnetic field, and m, = mass
of an individual electron [63]. From this equation, it is found that using a 2.45 GHz
excitation frequency, a static magnetic field of 875 Gauss is necessary for ECR to occur.
Based upon the experimental results from Wavemat and from theoretical predictions, the
limit for the ECR heating of the plasma-disk region of the MPDR-610 is 10 mTorr. At
pressures higher than 10 mTorr, a plasma is still present but collisional heating dominates.

The plasma chemistry is very complex for the oxygen system. 0‘ and 03 have
been detected in the laser-induced plume and are the believed mechanisms for
enhancement of laser-ablated and e-beam evaporated films at lowered molecular oxygen
partial pressures [31]. Recently, ion density data was determined for the MPDR-610,
ECR plasma generator [64]. Using a langmuir probe, initial measurements showed that
argon ion densities were above 10”/cm3 for 164 watts of incident power. Measurements
of argon ion density were as high as 1.4 x 10“/cm3 over the pressure range of .3 to .75
mTorr. Ion density measurement of our MPDR-610 oxygen plasma is not available since
it requires more complex measurement techniques such as laser induced fluorescence
(LIF). Others have reported atomic flux values for MPDR-610 oxygen plasmas using a
silver-coated quartz deposition monitor technique [65]. Values were as high as 9 x 10“
atoms/cmz-sec at 2 mTorr and 200 watts incident. Plasma density was also found to vary
downstream[64]. Hence, we designed three custom lengths of conflat full-nipples and used
a double-sided conflat flange to allow variable MPDR-to-substrate distances of 3.5 to 12

cm.

39
4.2.3 Optics

The excimer lasers used for ablation of YBCO operate at wavelengths of 351 nm
and as short as 193 nm. Fused silica windows and lenses met the requirement of high
transmission at wavelengths 2 193nm. A 30 cm focal length lens was mounted on a
gimbal outside the vacuum system. This allowed for greater versatility in spot size
adjustment and laser fluence.

The port which houses the silica window was designed to be 1.5" longer than the
rest of the ports. The additional length helped reduce the amount of evaporated YBCO
material deposited on the laser window. The evaporated material coats the windows,
resulting in optical losses. We cleaned the laser window after every four (depositions to
minimize the loss.

4.2.4 Target port

A rotatable, O-ring feedthrough was used instead of a bellows feedthrough to
mount and rotate the target. The advantage of the O—ring feedthrough is linear motion
toward or away from the substrate in addition to radial rotation. This allowed more
variance of target-to-substrate distance. The O-ring rotatable feedthrough was also
modified with bushings to reduce grinding of the shaft. A 60 rpm analog motor rotated
a gear set which reduced the target rotation to 16 rpm. This speed was sufficient to

ablate most of the target area evenly and reduce target pitting.
4.2.5 Vacuum Pump

A turbomolecular pump was chosen for its high pumping speed and pumping

capacity. Furthermore the speed of the pump could be varied so variable gas loads could

be handled. The pump was corrosion resistant so that Tulien pump oil could be used

40
pumping pure oxygen.
4.2.6 Substrate Heater

Blank et a1. studied effects of substrate temperature on T, of laser-ablated films

on SrTiO3 and MgO. Within the :40 °C region around 740 °C (see figure 4.3),
differences in T, were found to be only :5 K [66]. Other groups have found 760-780 °C
the ideal range for in situ c-axis growth of YBCO on YSZ [67,68,69]. At temperatures
below 500 °C, the lattice mismatch between film and substrate brings about strain; thus,
defects arise in the film which are deleterious to the superconducting properties [70]. The
strain depends on oxygen content since the c-axis length changes when the oxygen enters
the YBCO film[21]. This strain does not affect c-axis oriented films, but can be
deleterious to the quality of a-axis oriented films.

Thermocouple measurement of the actual substrate surface temperature is quite
difficult for thermal insulating substrates such as YSZ. They are also optically thin at the
IR wavelength used by pyrometer. When using a thermocouple to measure the surface
temperature of the substrate and comparing it to the heater block temperature it is found
that substrate is 50 - 150 °C cooler than the heater block temperature [71]. Lateral
variations in temperature across the heater block are found, leading to a temperature
uncertainty of :50 °C.

Because of the temperature difference between the block and substrate, the heater
block should be able to achieve temperatures in the range of 700 - 900 °C. It must also
be able to withstand the high-oxygen—pressure cool down procedure. A tungsten filament,
Si3N4 heater was used for deposition of YBCO in molecular oxygen. A copper clip was

designed to clamp the substrate to the Si3N,. A copper shield was designed which

41

 

     

 

 

 

 

90 E E ° I smor

g 80 - ------------ f- ---------------- : ------------
is" 5 s 5 m
g 70 ~ ------------ 'r -------- w: ------------ jr —————— — - -
m I i
a. I l
2 i I
LIJ 1 l
t— 50 — --------------------- 1 ------------ .- ----------- 1
_‘ I I
< i 1
5:2 . .
5g : :
o 50 ‘1 """"" ‘11 ----------- :- ........... J

40 ‘ 7 i i

600 650 700 750 800

SUBSTRATE TEMPERATURE (C)

Figure 4.3: The influence of substrate temperature on T,(R=0) for SrTiO,, MgO, and
Si + Zr02.(Afier Blank, Reference 66)

enclosed the heater for reduction of radiative heat loss. The shield enclosed most of the
heater except for a one-inch square window, exposing the substrate to the plume.

An applied voltage of 45 to 65 V was necessary to bring the substrate to the
desired temperature. Deposition of YBCO using the MPDR required a different heater
because the Si3N, heater shorted to ground (at an applied voltage 232 V AC) in the
presence of the highly-conductive oxygen-plasma. Therefore, a low-voltage resistive-

heater or optical-heater was necessary.

42

The selection of the heating filament for use in a resistive heater must be
compatible with the YBCO deposition process. For example, transition metals substitute
themselves for copper in the Cu-O chain and are detrimental to YBCO's superconducting
properties [72]. Researchers also discovered that molybdenum forms a volatile, high-
vapor-pressure oxide which affects formation of the orthorhombic, superconducting phase
of YBCO [22]. We suspect that rhenium also forms volatile oxides in the presence of
oxygen at high pressures. These oxides have similar contaminating effects on YBCO
formation. Most metals, (including many transitions) oxidize during deposition and
especially during the elevated oxygen partial-pressure (100-300 Torr), cool down
procedure; thus, the volatility of these oxides might lead to contamination even when the
metals themselves have low vapor pressure. A tantalum-filament, alumina body heater
was used for most depositions with the MPDR oxygen source.

4.3 YBCO Growth Process Design Considerations
4.3.1 Target-to-Substrate Distance

Studies have been performed to determine the fluence limit for cvaxis film growth.
1 to 3 J/cmz/pulse has been found to produce c-axis oriented YBCO films on various
substrates [51,53]. Since the laser-induced plume boundaries/characteristics change
drastically at low pressures (<40mTorr), fluence (as well as target-to-substrate distance)
must be re-optimized. It was found that two velocity components (of the
species/fragments emitted from the laser plume) exist and their time-of-flight (TOF)
distributions are affected drastically by oxygen partial pressure [69]. The TOF
distribution curves of atomic Ba in 10, 20, and 40 mTorr of oxygen indicate the larger

velocity component is weakened at 40 mTorr, and the smaller velocity components were

43

quenched at 10 and 20 mTorr. The plume region where the ablated atoms collide with
O2 is closely related to the quality of YBCO films and this region may be further away
from the target at pressures <40 mTorr. At lowered oxygen partial pressures (<20
mTorr), the larger component is not slowed very much by the oxygen and no luminous
front is present. Due to the longer mean-free path of the residual gas, the plume species
extend further in the radial directions and may increase film uniformity. These results
indicate that lower fluences and longer target-to-substrate distances may be necessary at
low oxygen partial pressures (<20 mTorr) to produce in-situ laser-ablated c-axis oriented
YBCO films.
4.3.2 Length of Deposition
Carim et al. have discovered optimum laser-ablated c-axis growth conditions on

LaAlO3 when the film thickness is kept below .4 um [73]. Yang et al. also discovered
reduced J, values in films thicker than .4 urn on YSZ [74]. Beyond the .4 um thickness,
the occurrence of 90° grain boundaries brings about a transition to grains with their c-axis
parallel to the surface. Deposition times were limited from 10 to 15 minutes and then
film thickness was measured with a profilometer to ensure that the deposition time was
not too long.

4.3.3 Cooling Procedure

Another important condition exists for successful deposition of YBCO films: the

high-pressure oxygen soak during controlled cool-down after the laser is shut off.
Because of the thermodynamic instability of YBaZCuJOI oxygen outdiffuses from the film
at the deposition temperatures (780 °C) [75]. During deposition, it is necessary to form

the tetragonal phase of 123 which has the approximate composition of YBaQCu3Ox where

44

X==6.0 to 6.2. This phase is non—superconducting but must be grown initially or the
orthorhombic-II phase will not be achievable in situ or through post-annealing. When the
laser is shut off, the heater temperature is kept constant until the oxygen pressure is
increased to 200 Torr or higher. The heater is then cooled slowly (rate varies from 1 hour
to several hours) to room temperature. During this time, the film soaks up oxygen until
the oxygen level in the films reaches the composition of X=6.93-6.98 [76]. The

thermodynamic stability limit of YBazCu307,x in molecular oxygen is given as

log PO (atrn) = 10.244 - 15,899/1‘
2
for decomposition by the reaction
3YzBaCuOS + 8BaCu202 + YBa3Cu206,y + n0Z '3 7YBa2Cu3O7,,,,

where n=6-(y+7x)/2 [77]. A diagram of the tetragonal, orthorhombic-I and orthorhombic-

 

    
   
   

  
   
   

 

 

 

103-
mommy Ortho-I Ortho-II
I Tetrogonol =-. 5,9
102-
16 y = 6.0 6.5
E? ,3
a ‘0 ‘5
2 I
o- :
3:, 1001
E i
Q. .
8 10-1, YZBaCu05 + 3001102 + Cu20
10-2 r 7 T I T l .
0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5

1000/1 [K"(-I)]

Figure 4.4: Oxygen partial pressure vs. temperature plot showing critical stability
limits for three phases of YBCO.(Afier Hammond, Reference 78)

45

11 phase pressure—stability limits are shown in figure 4.4 [78]. From the figure we can
find that the important condition to achieve proper stoichiometry is to stay above the
thermodynamic stability limit of YBaZCu3O7,X during deposition. Afterward, the film is
cooled at a rate sufficiently slow and pressure sufficiently high to move the oxygen

composition toward X=7.

Chapter 5
Results
5. 1 Introduction

The deposition parameters for YBCO films on MgO, SrTiO3, LaA103, and YSZ
are well established at molecular-oxygen partial-pressures of 150-200 mTorr and substrate
temperatures of 650-800 °C [48,79]. The processing parameters vary among ablation
systems due to differences in substrate heating apparatus, laser spots, focussing, and
cooling procedures. Optimization of our laser-ablation system using molecular oxygen
at 150-200 mTorr was an important first step toward finding the optimum parameter
ranges for successful deposition of YBCO films at lower pressures. The parameters to
be optimized for ideal, in-situ c-axis growth include: heater temperature, target-to-
substrate distance, and cooling procedure. The heater temperature is important for film
uniformity and for stable formation of the orthorhombic superconducting phase of YBCO.
Target-to—substrate distance and cooling procedure are directly related to crystal structure
and stoichiometry.

After successfirl deposition of YBCO films in molecular oxygen, activated oxygen
was incorporated using a tunable-stub microwave cavity and an MPDR oxygen source.
The next goal was to optimize the MPDR-610 oxygen source for use with the laser
ablation process and to study the effects of the MPDR on the quality of YBCO films
deposited at low pressures (<10 mTorr). The quality of YBCO films were monitored by
room-temperature resistance or R(300 K), R vs. T, J, vs. T, scanning electron

microscopy (SEM), electron microprobe X-ray emission, and X-ray diffraction.

46

47

5.2 Experimental Procedure for YBCO Depositions

The one-inch square polished (100) YSZ wafers were cut into various sizes.
Cleaning involved an initial ultrasonic bath in 2% Liquinox/Deionized (DI) water solution
for 15 minutes to remove organic impurities. Substrates were then rinsed in DI water for
five minutes and rinsed with isopropyl. A 15 minute ultrasonic bath in methanol was
then used to help remove any residual liquinox. Two isopropyl alcohol ultrasonic baths
were employed for removal of water before storage in isopropyl until use.

The deposition temperature was measured using a thermocouple and a pyrometer.
A K-type (chromel-alumel) thermocouple was clamped to the substrate surface and/or
heater block and read through the digital display of an temperature controller. The
pyrometer was focussed either on the film or the heater block during deposition. This
measurement depended on the emissivity setting. For our Si3N, heater, this value was
.71(emissivity of Si3N,) x .92(transmittance of type-708 glass at 7t=2 um) = 0.65.

The chamber was pumped down (using turbomolecular pump) to a base pressure
of <10" Torr before each deposition. The pressure could be controlled by sending the
0 - 10 V output of the baratron pressure gauge to feedback speed-control unit on the
turbopump power supply. During depositions in molecular oxygen, a needle valve
controlled the oxygen gas flow which entered the system 1 — 2 cm from the substrate via
an Va" copper tube. For depositions using the tunable—stub activated oxygen source, a
quartz flow-tube (coated with boric acid) was used to flow oxygen near the substrate.
During deposition with the MPDR oxygen source, oxygen was flowed through a small
capillary in the MPDR and the rate was controlled with a mass-flow controller. Flow

rates were in the range of .5 to 10 sccm. In addition, the oxygen was flowed into the

 

 

48

system after the substrate reached a temperature of 450 °C to avoid depletion of oxygen
on the surface of the YSZ substrates.

Three Lambda Physik variable-mode excimer-lasers with wavelengths of 193 nm
(ArF), 308 nm (XeCl), and 351 nm(XeF) were used. Pulse widths were 8 - 20 us at a
repetition rate of 5 Hz. The beam was focussed through a 30 cm-focal-length, fused-
silica lens onto the YBCO target. A laser fluence in the range of 1.6-8.7 mJ/cmz/pulse
was used, calculated from a 120-650 mJ/pulse focussed down to a (1 mm :t.2 mm) x (6
mm :2 mm) spot. The fluence calculation is only accurate to ~1 J/cmz. This is due to
the error in measurement of the spot dimensions, and the non-uniformity of the laser pulse
which produces regions of higher fluence across the length of the spot. The one-inch
stoichiometric target was rotated at approximately 16 rpm using an analog motor. Taking
into account that film thickness should be less than .4 pm to avoid outward a-b axis grain
misorientation, film depositions were limited to 8 to 15 minutes with typical deposition
rates between 0.4 - 1.7 A/pulse or correspondingly 2.0 - 8.3 A/sec.

For depositions using the MPDR, the distance from the 3 cm diameter plasma-disk
region to the substrate was varied from 3.5 to 12 cm. A one-kilowatt, 2.45 GHz Aztex
microwave power supply was used with the MPDR-610. The supply power was fed via
n-type, helicoflex cable to an isolator module which contains a dummy load and power
meters. This unit prevents reflected power from damaging the power supply. Another
section of helicoflex cable connects directly to the MPDR-610. The input power was 200
watts during deposition (above 200 watts, excessive heat degrades the rare-earth ECR
magnets which surround the plasma¢disk region). The percentage of input power coupled

to the plasma disk region was as high as 98%.

49

Substrate cooling consisted of increasing the oxygen partial pressure to 200 Torr
and then reducing the substrate temperature to approximately 450 °C in 30 minutes.
Pressure was then increased to one atmosphere and cooled to room temperature in an
hour. Several of our resistive heaters (which were used in conjunction with the MPDR)
did not permit slow cooling and films were quenched in oxygen as the heaters shorted out
during the high-pressure cool-down.

5.3 Results of YBCO Growth Optimization
5.3.1 Heater Temperature

Initially, we suspected that heater temperature was the most critical parameter for
successful c-axis growth. Laser fluence and target distance were thought to be less
critical due to severe fluctuations in room temperature resistance, R(300 K), of our films
with small changes in heater temperature (110° C). These films were non-superconducting
as-deposited. The actual substrate temperature was only accurate to within 50-160°C
because of the difference in substrate temperature as compared to the heater block
temperature. We used a K—type thermocouple (accuracy of this measurement varies also
due to difference between thermocouple junction temperature and substrate surface
temperature) and a pyrometer to measure the substrate temperature and the heater-block
temperature. Our measurements indicate that the substrate temperature was measured to
be as much as 160°C lower than the block temperature.

Various heaters were employed including a tungsten-filament Si3N4 heater, a
graphite-filament alumina heater, a tantalumcfilament alumina heater, and a halogen-lamp
(tungsten-filament) optical heater. The Si;,N4 substrate heater was successfully used for

deposition in 200 mTorr of molecular oxygen to produce YBCO films with T,(R=0) as

50
high as 88.5 K. The high heater potential of approximately 70 V needed to achieve a

temperature of 780 - 800°C posed a significant setback when depositing YBCO in
conjunction with the MPDR-610 oxygen source. From our experiments, it was discovered
that the upper limit of the heater potential for use with the MPDR was 32 V. Beyond 32
V, undesirable arcing occurs through the highly conductive plasma. The arcing causes
the heater leads to short out to ground and may damage to the quartz jar surrounding the
plasma disk region. Also, evaporation of metal (possibly due to the arcing problem) may
have caused migration of transition metals into the YBCO films.

Low-voltage, resistive heaters using platinum wire, rhenium ribbon, and graphite
foil were all used in attempts to reduce the heater potential. Platinum wire shorted out
during deposition. The graphite filament heater was used since it required only 12 V to
drive it to a temperature of 800 °C. Unfortunately, based upon our R vs. T measurements
and our findings in the literature, we suspect that non-conducting surface-layers form in
the presence of CO2 [80] and degrade film quality. We also suspect the rhenium ribbon
formed volatile oxides which were detrimental to the superconducting properties. Finally,
we used tantalum wire and a tantalum plate on a resistive-filament, alumina heater. The
tantalum filament heater reduced contamination of the surface, but did not perform well
during the high oxygen partial pressure cool-down. Films can be oxygen deficient
without proper cooling in oxygen, and we suspect heater bum-out may be responsible for
lowered 'I‘c values.

5.3.2 Target—to-Substrate Distance
By switching from the 'constant voltage' mode to the 'constant energy' mode on

the excimer laser, we found the plume shape/size became more uniform. In the constant

51

voltage mode the energy per pulse varied as much as :40 ml as compared to :4 m].
Soon after (#LA-25, Tc=89 K, high degree of c-axis orientation), we discovered the
optimum growth condition occurred where the substrate was positioned at the visible edge
of the plume. This finding is supported by recent studies which correlate the luminous
tip of the laser plume with optimum c-axis YBCO film growth [53,57]. This distance was
in the range of 4 to 6 cm, depending on the laser fluence and oxygen partial pressure.
5.3.3 MPDR-610 Oxygen Source

Since the MPDR-610 ion/free radical source was one of the first produced, we
encountered several problems: incorrect factory tuning/operating specifications and the
presence of conducting media in front of the plasma disk region. The center-conductor
and sliding short positions on the MPDR-610 (specified by Wavemat) did not result in
successful operation of the source. Using different lengths of 4.5" confiat full-nipples,
the tuning of the MPDR cavity was optimized at various downstream distances from the
heater and at various pressures (see figure 5.1). As the figure shows, even at the
minimum cavity tuning conditions, (shown here at 8 mTorr) ~85% of the incident power
is coupled to the plasma disk region. The source required operating pressures S4 mTorr
to achieve the highest power coupling. This pressure limit (4 mTorr) also corresponded
to a visually brighter plasma. At pressures from 4 to 8 mTorr, reflected-power from the
plasma—disk region increases and shorter mean-free path of residual gas restrict 0' flux
at or near the substrate.

These results are important since electron cyclotron resonance (ECR) heating is
a low-pressure coupling technique and the MPDR-610 has a useful ECR operating range

from .1 to 10 mTorr. The closest possible MPDR to substrate distance in our laser

[for.POWERmx]/[ref.POWER]

10

1O

52

 

+ MPDR 5 cm from substrate heater
x MPDR 7 cm from substrate heater
“ MPDR 12.5 cm from substrate heate
a MPDR 15 cm from substrate heater

“j .

 

 

 

3‘ an
i" 1 i 3‘ x
fir; 0:30 firs g g § § 5 +
x
§
*3
i
+
leTjr—l—rjjllT—T]filrrTlTllll—rrjljll
1 6 11 16 21 26 31 3

PRESSURE (mTorr)

Figure 5.1: Ratio of maximum forward power to reflected power of the MPDR-610
vs. oxygen partial pressure. The substrate heater is placed at various distances from
the MPDR. The input power is 200 Watts of 2.45 GHz microwave radiation.

6

 

53

ablation system is 3.5 cm. At 10 mTorr, the mean free path of residual gas is 4.2 cm.
This limits the useful operating pressure of the MPDR-610 in conjunction with the laser
ablation system. If the substrate is placed too far away from the plasma-disk region, the
atomic flux near the substrate may be too low to enhance oxygen composition of the
YBCO films. We found that the closest possible MPDR-to-substrate distance (3.5 cm)
and 200 watts incident resulted in the highest quality YBCO films.

5.4 Deposition of YBCO films in Molecular Oxygen

After optimization, successful deposition of YBCO films at high (180-200 mTorr)
molecular oxygen partial pressures was achieved at all laser wavelengths used: 193, 308,
and 351 um. All depositions were performed using 5 Hz laser pulse fired on a target
rotating at 16 rpm, with a typical laser fluence of 1.6 - 8.7 J/cmz/pulse. Most YSZ
substrate temperatures were in the range of 720-820 °C. The chamber pressure was
pumped below 10" Torr before flowing oxygen near the substrate.

Table 5.1 highlights some of the films deposited using molecular oxygen. Critical
temperatures, TC(R=0), were as high as 89 K (sample #LA-25) on yttria-stabilized zirconia
(YSZ). Samples exhibited narrow R=0 transitions, as evidenced by figure 5.2. Resistance
vs. temperature plots were made using the AC 4-point probe measurement technique.
Contact to the film was made by applying pressure. These results are comparable to other
groups whose TC(R=0)'s as high as 92 K [81]. Jc measurements were made on film
#CLA-Z (200 mTorr Oz) by patterning a 40 um wide bridge using wet-etching and
photolithography. A diagram of the bridge is shown in figure 5.3. Contact was made by

exerting pressure on the four contacts. The bridge critical current was measured by

ramping a nanosecond current pulse through the outer two contacts. Monitoring the

54

Table 5.1 YBCO films with TC(R=O) between 73 and 89 K. Films were deposited
using molecular oxygen.

Subs:rate/ Laser Laser
Sample Heater Pressure Wavelength Energy R(3OOK1
Number Temperature (mTorr) A (mJ) (0)
(°C) (am)

OLA-82
ILA-78
ICLA-dS
ILA-56
QCLA-B
OLA-76
ILA-S9
0LA-Sl
ILA-54
iLA-la
ILA-60
OLA-l7
OLA-58
fLA-ZS
OCLA-Z
OLA-26

OLA-29

 

55

 

1.0
0.9 ~
0.8 4
0.7 ~
0.6 a
0.5 4
0.4 .1
0.3 a
0.2 ..
0.1 e

R (K) / R (300K)

 

 

#CLAZB, YBCO an YSZ,
200 mTorr 02

 

 

0.0 TWWijTl

0

I‘lllllfiIWTTIVl—lflTl—l—r

100 200
TEMPERATURE (K)

Figure 5.2: Resistance vs. Temperature for YBCO on (100) YSZ. The substrate

temperature was 780 °C.

300

 

56

 

Figure 5.3: Diagram of the patterned YBCO microbridge. The I, V designations
represent the 4-point probe contacts.

57

voltage ramp across V+ and V', the point along the ramp at which the bridge goes into
the normal state, represents the bridge critical current.

The measured Jc value at 78.5 K was approximately 1.6 x 106 A/cm2 (figure 5.4)

for sample #CLA-Z, but because the bridge burned out from the high current pulse, a
measurement at 77 K is not available. Extrapolating the Jc curve, the value at 77 K is
approximately 1.9 x 10‘ A/cm2 which is comparable to several other results (Tc-:92 K, 200
mTorr 02) of up to 5 x 106 A/cm2 at 77 K[68]. The bridge is wide enough to ignore the
mask defined cross-sectional error (MDCS) found with smaller linewidth bridges. The
Jc measurement is still only accurate to a factor of 2 because of variance in bridge height
(figure 5.5) across its length. This height variance is due to non-uniform thickness of
films produced at 100-200 mTorr and the existence of surface particles.

Figure 5.6 shows the plume geometry which has a strong effect on film thickness.
As the figure shows, the elliptical shape of the plume can cause film thickness to vary
dramatically from the center to the angles away from the target surface normal. Our
measurements by profilometer across a 1" YSZ wafers show the film is less than 1/2 the
thickness at the edge as compared to the thickness at target-surface-normal.

Electron microprobe X-ray emission was used to measure relative stoichiometry
over a 10 mm vertical distance. Figure 5.7 shows the film has the 123 stoichiometry over
most of the measured vertical positions. Over a 6 mm scan region, the stoichiometry is
very close to '123'. At the edges of the sample, the stoichiometry is non-uniform and
does not have the 123 composition.

X-ray diffraction was used to determine crystalline quality. Figure 5.8 shows the

Bragg diffi'action peaks of the planes parallel to the c-axis, as scanned over 26) angles of

58

 

2.2
2.0 -

1'8 ‘ #cuue. YBCO on YSZ
15 _ 200mTorr Molecular oxygen
' Tc(R=0)=88.5 K

1.4 '1
1.2 -
1.0 -
0.8 .
0.6 -
0.4 —

0.2 T l l l l T l
77 78 79 80 81 82 83 84 85

Temperature (K)

Jc (millions A/cm2)

 

 

 

 

Figure 5.4: Critical Current (1,) vs. Temperature for YBCO film sample #CLA-2.

59

 

9., i " 6,000
5,000

 

 

‘ Y‘V VV‘V'V

 

-, -w ; 3,000
- 2,000
1,000

 

 

 

4 .'

 

’ ~
D
r

 

 

 

’ ‘ 1 ._ 4000

t

1

 

 

 

 

 

 

 

 

 

 

 

400 420 440 460 480 500 520 540
Sean Distance (um)

Figure 5.5: Profilometer measurement of a 40 um wide YBCO bridge.

60

Target Surface Normal

% m <5 Substrate
/ \

 

 

 

 

Figure 5.6: Plume Geometry in the laser ablation deposition system.

01

Relative stoichiometry
_. m c.) A

O
0

§

._,__. _.__ -_-.... - i

Sample 1. Cu/Y
Sample 2. Cu/Y

61

 

l 1 L
6 e 10
position on film, mm.

+ Sample 1. Ba/Y
' Sample 2. Ba/Y

12

Figure 5.7: Relative stoichiometry vs. position for YBCO film deposited in molecular

oxygen.

  
   

Intensity(arb. units)

” i",:
, iif'.‘

26 (degreeS)

Figure 5.8: X-ray diffraction pattern of YBCO film on (100) YSZ.

63

6 to 60°. The X-ray pattern of this sample revealed the films had a high-degree of c-axis
orientation as evidenced by the well-defined Bragg diffraction peaks of the (001) through
(009) planes.

Scanning electron micrOSCOpy (SEM) showed some films had a granular surface
morphology and some films were quite smooth (figure 5.9). All films had surface
particles 1 pm or larger in size, which are quite common in laser ablated films [47,82].
It has been shown that their presence does not impede Jc or Tc values, but their size and
frequency of occurrence are affected by laser fluence and wavelength. The smaller the
wavelength and lower the fluence, the smaller and fewer the particles. Rapid Thermal
annealing for up to 5 minutes at temperatures as high as 920°C does not eliminate these
particles. The granular surface morphology is consistent with films made by Yang et a1.
[74], and Ramesh et al have identified the origin of the grainy outgrowths[83]. The
lenticular—shaped outgrowths are caused by a,b-axis regions which nucleate
heterogeneously at second phase regions. This second phase has been identified as the
132 phase or YBa3CuZO7,x.

5.5 Deposition of YBCO Films Using a Microwave Cavity

Using similar growth conditions to films made between 100—200 mTorr 02, the
effect of atomic oxygen from a tunable-stub microwave cavity was investigated. A
similar apparatus to the one used by Greer [29] was employed here using a 1/2" tube
tapered down to a 2 mm diameter nozzle near the substrate. The quartz flow-tube (coated
with boric acid to reduce recombination of activated oxygen species) was directed at the
substrate from a distance of 2 cm. Pressures were varied from 30 to 185 mTorr. Sample

#LA-73, prepared at 70 mTorr and 785°C, exhibited a TC(R=O) of 86.5 K (figure 5.10)

 

1 H“: ZUKU

Figure 5.9: Scanning electron micrograph (SEM) images of the surface of a YBCO
film on YSZ.

65

 

1.0
0.9 a
0.8 _
0.7 a
0.6 4
0.5 -
0.4 -
0.3 '1
0.2 a
0.1 ~

  

#LLA73, 70 mTorr

 

Resistance (ohms)

 

 

 

 

0.0 lTlTfiTlrlrlrWTTlrj T'Tlrllrllrj

0 100 200
Temperature (K)

Figure 5.10: Resistance vs. Temperature of YBCO film on (100) YSZ deposited
using a microwave cavity oxygen source.

300

66

and J,, was measured to be 1.1 x 106 A/cm2 at 77 K (figure 5.11). Table 5.2 summarizes
the comparisons between samples prepared in molecular oxygen and samples prepared in
microwave activated oxygen. The results are inconsistent as samples prepared at 75 mTorr
show improvements with atomic oxygen (see figure 5.12), while other samples prepared
in 02 exhibit TC(R=0)'s slightly higher. Figure 5.12 shows the activated oxygen prepared
film has a TC(R==0) improvement of 5 K over the film prepared in molecular oxygen.
Table 5.2 shows that the activated-oxygen sample prepared at 35 mTorr has a TC(R=O)
which measures 0.5 K lower than the sample prepared in molecular oxygen at the same
pressure.

As discussed earlier, the plume shape changes at lower pressures (<40mTorr) due
to the longer mean-free path of residual gas molecules. Optimizing substrate position for
premium in-situ c-axis growth is currently under investigation. The 180—200 mTorr
activated-oxygen assisted film quality results are in accordance with Fork et a1. [68].
They found that the use of a 2.45 GHz atomic oxygen source did not appear to improve
film quality on sapphire substrates. The effect of the microwave-cavity activated oxygen
source at this pressure range (35-185 mTorr), may have had little or no effect on YBCO
film quality.

5.6 Deposition of YBCO Films Using an MPDR

Using the MPDR-610, sample #CLA-l3l (prepared at 815°C) shows low room
temperature resistance and exhibit R=0 at 79 K(see figure 5.13). The zero-resistance
transition is sharp, but a 'tail' exists at 83 K and the final superconducting state is not
achieved until 79 K. Sample #CLA-13l was deposited on YSZ at an oxygen partial

pressure of 4 mTorr, which to our knowledge is the best result in the literature to date.

67

 

   

#LLA73, YBCO on YSZ
Tc(R=0)=86.5 K

[\D
O
r

1.8 a
1.6 4
1.4 ..
1.2 r
1.0 ~
0.8

Jc (millions A/cm2)

 

 

 

l l r T l T 3 1 r r T
67 68 69 70 71 72 73 74 75 76 77 78 79
Temperature (K)

Figure 5.11: Critical Current (1,) vs. Temperature for YBCO sample #LLA-73. The
Jc value at 77 K is 1.1 x 10‘ A/cmz.

68

Table 5.2: Comparison of YBCO films deposited in molecular oxygen and films
deposited in activated oxygen from a tunable-stub microwave cavity.

 

 

 

 

 

 

 

 

 

 

 

Deposition Condrtions Measurements
Substrate/ Laser Laser
Sample Heater Pressure Wavelength Energy R(300K) TciOl
Number Temperature (mTorr) A (mJ) 10) (K)
(°C) (nm)
OLA—80 790 35 193 50 68.5
OLA-79 795 35' 193 50 68
OLA-82 800 70 193 40 83
ILA-78 800 75 193 275 60 76
OLA-73 785 70' 193 40 86.5
ILA-81 ? 70' 193 40 83
OLA-83 ? 70' 193 45 78.5
OLA—84 800 70' 193 40 84
OLA-77 79S 72' 193 290 50 81
ILA-75 795 73' 193 300 50 81

 

 

 

 

 

 

 

 

 

 

' Indicates tunable-stub microwave cavity oxygen source

69

 

10
09 ~
013~
07'-
0154
Ofiie
0A~~
0.38
01!-
OJ -

R(K) / R(300 K)

(a)

(b)

 

 

01)
0

Vi VI VT T—TTTj T rn T‘FW l fjfi—ivrjj l l l—T

100 200
Temperature (K)

 

300

Figure 5.121Normalized resistance vs. T (K) for two samples prepared in (a) 02 at 75
mTorr, TC(R=0)=7 6K and (b) 75 mTorr activated oxygen, T,(R=0)=81K.

70

 

 

 

 

 

200
#CLA131, Tc=79 K, 4 mTorr
70‘
E
.C
3
8 1004
C
O
.75
(I)
Q,
m
0TrTrrrfirrrfirrTWlfiFTFTT“l‘1‘l
0 100 200 300

Temperature (K)

Figure 5.13: Resistance vs. temperature for YBCO on YSZ. The film was grown
using the MPDR-610 oxygen source at 180 watts, 4mTorr, and Rm = 815°C.

71
The sample #CLA-13l is approximately 1800A thick and was patterned for J,

measurement. Figure 5.14 shows the plot of J, vs. temperature and the value at 77 K is
2.9 x 10’ A/cmz. This value is consistent with our own data and values in the literature.
For example, the measured J, value of sample #LLA-81 (T, = 83 K, 70 mTorr, YSZ) is
4.6 x 10’ A/cm2 which is within a factor of two of the #CLA-13l J, value. Berezin et
al. [84] reports a measured J, of 0.8 x 10’ Mom2 for a film grown by plasma—assisted e-
beam evaporation (T, = 80 K, ~10‘3 Torr, SrTiOJ).

The x-ray diffraction pattern of figure 5.15 shows that the film has c-axis
orientation. Figure 5.16 shows SEM images of the rougher surface morphology than
films prepared at higher pressures (200 mTorr).

A film deposited under similar conditions using molecular oxygen (4 mTorr, 780-
800° C), shows completely semiconductive behavior as a function of temperature (figure
5.17). Other films deposited using an MPDR at low pressures (.5 to 10 mTorr), exhibited
broad superconducting transitions or semiconductive behavior.

A possible explanation for the lowered T,(R=0) and broad R=0 transitions of some
of our films may lie in the nature of the oxygen-indiffusion cool-down step. For example,

our films prepared in molecular oxygen with the Si3N, heater were cooled slowly after

deposition. The heater was held at the deposition temperature until the chamber was

filled to 200 Torr of oxygen. The heater was then reduced to 450 °C and cooled to room
temperature in about an hour. This procedure consistently yielded films with T,(R=O)
values above 77 K and some were as high as 89 K.

Films which are oxygen deficient are usually characterized by broad R=0

transitions. These YBCO films may contain the orthorhomic-I phase which is

Jc (x 10‘6 A/cm2)

72

 

0.37
0.36 ~
0.35 -
0.34 1
0.33 1
0.32 -1
0.31 1
0.30 —
0.29 a
0.28 ‘

 
   

#CLA131, Tc=79 K, 4 mTorr

 

0.27

10

 

20

r

30 40 50 60 70
Temperature (K)

Figure 5.14: Critical current density (1,) vs. temperature for sample #CLA-131
prepared using the MPDR -610. The J, value at 77 K is 2.9 x 105 A/cmz.

 

73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20 (degrees)

Figure 5.15: X-ray diffraction pattern showing c-axis orientation of YBCO on (100)
YSZ The sample number rs CLA-131, prepared usrng the MPDR-610 at 4 mTorr

74

 

(b)

Figure 5.16: Scanning electron micrographs of film #CLA-

(b) l um resolutions.

131 at (a) 2 pm and

75

 

 

 

 

3000
1
1
g 1
I 2000:
8 .
LIJ
‘2’ 1
<
t— .
g} .
{a 1000 j
m
#CLA47, YBCO on YSZ,
780-800 C, 4 mTorr 02
0 l I l l I l f F I I T l l l T l l l l I l l l I T l l l T
0 100 200 300
TEMPERATURE (K)

Figure 5.17: Resistance vs. Temperature curve for YBCO on YSZ deposited at 4
mTorr Oz. The film exhibits semiconductive behavior.

76

superconducting at lower temperatures (<50-60 K) and has an oxygen content of in the
range of X = 6.2 to 6.5. This deficiency may be due in part to rapid cooling (quenching)
in conjunction with insufficient oxygen partial pressures. Researchers have determined
rapid quenching can cause an elongation of the c-axis (by approximately .2 A)[21]. Post-
deposition annealing can improve the T,(R=0), but in situ deposition is desired for
reduced film/substrate interaction and higher J, values. The burnout of the heater
filament used with MPDR depositions causes rapid cooling at approximately 2 Torr, well
below the 200 Torr oxygen partial pressure used with the Si3N, heater. With the
difficulty in devising an adequate heating apparatus (low-voltage, resistant to oxygen
soak) for use with the MPDR-610, inappropriate cool—down procedure may have been

responsible for the low T,(R=0) of these films.

Chapter 6
Summary and Conclusions
6.1 Summary

A laser ablation system which incorporates a microwave plasma disk reactor
oxygen source was designed and constructed. We successfully optimized the laser
ablation system for growth of YBCO films in molecular oxygen at high oxygen partial
pressures (>3 5mTorr). Best results were obtained at substrate temperatures of 750 to 820
°C with the luminous tip of the laser-induced plume at the substrate surface. For this
'luminous tip' condition, the target-to-substrate distancewas found to be 4 to 6 cm,
depending on laser fluence and oxygen partial pressure. T,(R=0) values were as high as
89 K and the measured J, at 78.5 K was as high as 1.6 x 10‘ A/cmz. Extrapolating the
J, curve, the value at 77 K was approximately 1.9 x 10‘ A/cmz. The surface morphology
was rough; it contained particles and lenticular-shaped outgrowths. X-ray diffraction
patterns showed a high degree of c—axis orientation.

A tunable-stub microwave cavity (excited using 2.45 OHz microwave radiation)
was used at high oxygen partial pressures (>35 mTorr) to deposit YBCO films. The
T,(R=0) was found to be as high as 86.5 K and J, was as high as 1.1 x 10‘ A/cm2 at 77
K. Activated oxygen enhanced films did not exhibit markedly better properties than
molecular oxygen films at these pressures.

Atomic-oxygen-enhanced laser-ablation deposition at low-pressures (<_lOmTorr)
was employed using an MPDR at an excitation frequency of 2.45 GHz. The ECR oxygen

plasma from the MPDR created a high atomic-oxygen flux region (approx. 10” /cm2-sec)

77

78
and permitted successful deposition of YBCO on YSZ at 4 mTorr. The T,(R=0) was 79

K for the best film, and the J, for the same film measured 2.9 x 10‘ A/cm2 at 77 K. X-
ray diffraction patterns revealed that the film was c-axis oriented. SEM images showed
a much rougher surface morphology than films prepared at higher pressures (150-200
mTorr).

6.2 Conclusions and Suggestions

A possible explanation for the broad R=0 transitions of some of our films may lie
in the nature of the oxygen-indiffusion cool-down step. This deficiency may be due in
part to rapid cooling (quenching) in conjunction with insufficient oxygen partial pressures.
The burnout of the heater filament caused rapid cooling at approximately 2 Torr, well
below the 200 Torr oxygen partial pressure used with the Si3N4 heater. With the
difficulty in devising an adequate heating apparatus for use with the MPDR-610,
inappropriate cool-down procedure may have been responsible for the low T,(R=0) of
most of these films. Metal-oxide contamination may also have contributed to the poor
quality of these films.

Several modifications to the laser ablation apparatus may improve film quality at
low pressures. With an external optical-heater, better control of the cooling procedure
would be possible. An optical heater would be compatible with the MPDR and would
not introduce contaminants into the ablation chamber. A larger diameter processing
chamber may also help to optimize the target-to-substrate distance at low pressures.

Further work necessary for optimizing laser ablated YBCO growth at lowered
temperatures (<650°C) and at lowered oxygen partial pressures (<35 mTorr). One

suggestion is to use an excitation frequency of 915 MHz to form the ECR plasma.

79

Recent measurements indicate the atomic species present in the oxygen plasma of the

MPDR-610 are orders of magnitude higher for this excitation frequency [85].

LIST OF REFERENCES

A. Sleight, "Synthesis of Oxide Superconductors," Physics Today, June 1991,

pp.24-30.
"An Interview with Venky Venkatesan," Supercurrents, July 1989, 49-58.

[1]

[21
[3] R. Singh, J. Narayan, A. Singh, and J. Krishnaswamy, "In Situ Processing of
Epitaxial Y-Ba-Cu-O high T, Superconducting Films on (100) SrTiO3 and (100)
YS-ZrO2 Substrates at SOD-650°C," Appl. Phys. Lett. 54 (22), May 29, 1989,
pp.227l-3.
Kamo, and S. Matsuda, "Properties of

[4] T. Nabatame, Y. Saito, K. Aihara, T.
leBa2CaQCu3O, Thin Films with a Critical Temperature of 122 K Prepared by

Exicimer Laser Ablation," Jap. Jour. of App. Phys, V. 29, No. 10, Oct. 1990,

L18 13-1815.

[5] M. Dilorio, "Preparation of High-T, YBaQCu3O7,x Thin F ilms," Proc. of the SPIE
Conf., Vol. 1287, San Diego, March 20-21, 1990.

[6] Kittel, C., Introduction to Solid State Physics, Chapter 12, 1974.

J. File and R. G. Mills, Physical Rev. Lett. 10, 93, 1963.

l 71

[8] M. Aslam, Lecture Notes

[9] B. D. Josephson, Phys. Rev. Lett., 1, 251(1962).

[10] B. D. Josephson, Advan. Phys, 14, 419(1965).

[11] B. D. Josephson, Rev. Mod. Phys. 46, 251(1974).

[12] V. Ambegaokra and Baratoff, Phys. Rev. Lett, 10, 486,( 1963).
PIE 1 8 onfer nce n A vance in i n uct r and

" NeWport Beach, California,

[13] 1 C. Vanneste,
Supercgndugtors, ”Superconductor/Light Interaction,
March 14, 1988, pp 50-65.

[14] M. Decroux, "High T, Superconducting Based Sensors, " Sensors and Actuators,
A21 -A23(1 990)9-14.

80

81

[15] Robins, D., Introduction to Superconductivig, Chapter 3, IBC Technical Services

Ltd., London, 1989.

M. Beasley, "High Temperature Superconductive Thin Films," Proceedings of the

IEEE, V. 77, N.8, Aug.,l989, pp. 1155-1163.

[17] R. Roas, B. Hensel, G. Endres, L.Schultz, S. Klaumunzer, and G. Saemann -
Ischenko, "Superconducting Properties and Irradiation Induced Effects of Epitaxial

YBaCuO Thin Films,"Physica C, 162-164,(1989),135-6.

[16]

[18] D. Miller, P. Richards, S. Etemad, A. Inam, T. Venkatesan, B. Dutta, and X. Wu
"Residual Losses in Epitaxial Thin Films of YBa2Cu3O7 from Microwave to
Submillimeter Wave F requencies," Appl. Phys. Lett. 59 (18), Oct. 28,1991 ,2326-8.

[19] L. Lee, K. Char, M. Colclough, and G. Zaharchuk, "Monolithic 77 K DC SQUID
Magnetometer," Appl. Phys. Lett. 59 (23), Dec. 2, 1991, 3051-3.

[20] R. Muenchausen, X. D. Wu, S. Foltyn, R. Estler, R. Dye, A. Garcia, and N
Nogar, "High Rate Growth of YBa2Cu307,x Thin Films Using Pulsed Excimer
Laser Deposition," Mat. Res. Symp. Proc. V191, 177-181.

[21] T. Goto, H. Masumoto, and T. Hirai, "Preparation of YBazCu3O,_x Films by ECR
Plasma Sputtering," J.Journ. of Appl. Phys, V. 28, (1), 1989, pp.L88-90.

[22] D. Kubinski and D. Hoffman, "Reactive Codeposition of In Situ Y-Ba-Cu-O
Superconducting Films Using Dilute Mixtures of Ozone in Oxygen," J. Appl.
Phys.,V 71(4), Feb. 1992. PP.1860-7.

[23] R. Singh and J. Narayan, "Pulsed-Laser Evaporation Technique for Deposition of
Thin Films: Physics and Theoretical Model," Physical Rev. B, VA], (13), Mayl,

1 990,8843-59.
[24] J. Dawson, P. Kaw, and B. Green, Phys. Fluids 12, 875 (1969).

[25] S. Witanachchi, H. Kwok, X. Wang, and D. Shaw, "Deposition of Superconducting
Y-Ba-Cu-O Films at 400°C Without Post-annealing," Appl. Phys. Lett., 53 (3),
July 18, 1988, pp. 234-6.

[26] H. Izumi, K. Ohata, T. Hase, K. Suzuki, T. Morishita, and S. Tanaka,
"Superconductivity and Crystallinity of BazYCu;,O7 x Thin Films Prepared by
Pulsed Laser Deposition with Substrate Bias Voltage," J. Appl. Phys, 68( 12) Dec.
15, 1990, 6331-35.

[27] A. Gupta and B. Hussey, "Laser Deposition of YBa2Cu3O.,,x Films Using a Pulsed
Oxygen Source," Appl. Phys. Lett. 58 (11), Mr. 18, 1991, pp. 1211-3.

82

[28] J. Kwo, M. Hong, D. Trevor, R. Fleming, A. White, J. Mannaerts, R. Farrow, A.
Kortan, and K. Short, "In Situ Growth of YBazCu3O7,x Films by Molecular Beam
Epitaxy with an Activated Oxygen Source," Physica C 162- 164( l989),pp. 623-624.

[29] J. Greer, "In-Situ Growth of YBaZCu3O7_x Thin Films on Three-Inch Wafers Using
Laser-Ablation and an Atomic Oxygen Source," Proceedings of the Third Annual

Conference on Superconductivity and Applications, Buffalo, N. Y., Sept. 1989.

[30] K. Yamamoto, B. Lairson, C. Eom, R. Hammond, J. Bravman, and T. Geballe,
"Role of Atomic Oxygen Produced by an Electron Cyclotron Resonance Plasma
in the Oxidation of YBa2Cu3O7,x Thin Films Studied by In Situ Resistivity
Measurement," Appl. Phys. Lett. 57 (18), Oct. 29, l990,pp. 1936-1938.

[31] R. Humphreys, N. Chew, J. Edwards,J. Satchell, S. Goodyear, and S. Blenkinsop,
"Oxygenation of YBaZCuJOx Thin Films and Multilayers," Supercond. Sci.
Technol. 4 (1991), pp. 3172-174.

[32] P. O’Keeffe, S. Komuro, S. Den, T. Morikawa, and Y. Aoyagi, "The Irradiation
Effects of an Oxygen Radical Beam on the Preparation of Superconducting Thin
Films," Jap. Jour. of Appl. Phys. V. 30 (5A), May, 1991, pp. L834-L837.

[33] P. O'Keeffe, S. Komuro, S. Den, T. Moriwawa, and Y. Aoyagi, "Development and
Applications of a Compact Electron Cyclotron Resonance Source," Jap. Jour. of

Appl. Phys. v. 30 (1113), Nov., 1991, pp. 3164-3168.

[34] M. Norton and C. Carter, "Effect of Substrate on the Early Stages of the Growth
of YBaQCu307,x Thin-Films," Mat. Res. Soc. Symp. Proc. V. 191 ,1990,pp.165-l70.

[3 5] R. Singh and Narayan, "Nature of Epitaxial Growth of High-T, laser-deposited Y-
Ba-Cu-O Films on (100) Strontium Titanate Substrates," J. Appl. Phys. 67(8), Apr.

15, 1990, 3785-90.

[36] H.0hlsen, J. Huddner, and L. Stolt, " Thin Films of Y-Ba-Cu-O on Sapphire
Substrates," Proc. of ICMC'90 Top. Conf. on High Temperature Superconductor

Material Aspects, May 9-11, 1990.
[37] H. Ohlsen, J. Hudner, L. Stolt, and E. Johansson, "Y-Ba-Cu-O Thin Films on
Sapphire Substrates," Physica C 162-164 (1989), pp. 621-622.

[38] P. Tiwari, S. Sharan, and J. Narayan, "In Situ Single-Chamber Laser Processing
of YBa1,Cu,O,,x Superconducting Thin F ilrns on Yttria-Stabilized Zirconia Buffered

(100) GaAs," Appl. Phys. Lett., 59 (3), July 15, 1991, pp.357-9.

[39] M. Rao, E. Tarsa, L. Samoska, J. English, A. Gossard, H. Kroemer, P. Petroff,E
Hu, "Superconducting YBaCuO Thin F ilrns on GaAs/AlGaAs," Appl. Phys. Lett.

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

83
56 (19), May 7, 1990, pp. 1905-7.

X. Wu, R. Muenchausen, N. Nogar, A. Pique, R. Edwards, B. Wilkens, T. Ravi,
D. Hwang, and C.Chen, "Epitaxial Yttria-Stabilized Zirconia on (1102) Sapphire
for YBazCu307,x Thin Films," Appl. Phys. Lett. 58 (3), jan. 21, 1991, pp. 304-6.

M. Aslam, R. Soltis, E. Logothetis, R. Ager, M. Mikkor, W. Win, J. Chen, and
L. Wenger, "Rapid Thermal Annealing of YBaCuO Films on Si and SiO2
Substrates," App. Phys. Lett. 53 (2), July 11, 1988, pp. 153-155.

A. Fartash, I. Schuller, and J. Pearson, "Solid-State Reactions in High-
Temperature Superconductor-Ceramic Interfaces: Y-Ba-Cu-O on A1203 versus
Yttria-Stabilized ZrOz, and MgO," J. Appl. Phys.,67(5), Mar. l,1990,pp.2524-2527.

G. Samara, "Low-Temperature Dielectric Properties of Candidate Substrates for
High-Temperature Superconductors: LaAlO3 and ZrOzz9.5 mol % YZO3," J. Appl.
Phys, 68 (8), Oxt. 15, 1990, pp. 4214-4219.

H.Izumi, K. Ohata, T. Sawada, t. Morishita,a nd S. Tanaka, "Deposition Pressure
Effects on the Laser Plume of YBazCu3O,,x," Jap. Journ. of Appl. Phys, V. 30,
No. 9A, Sept, 1991, pp. 1956-58.

J. P. Zheng, Z. Huang, D. Shaw, H. Kwok, "Generation of High-Energy Atomic
Beams in Laser-Superconducting Target Interactions," Appl. Phys. Lett. 54 (3),
Jan. 16, 1989, pp. 280-2

K. Erington and N. Ianno, "Thin Films of Uniform Thickness by Pulsed Laser
Deposition," Mat. Res. Soc. Symp. Proc., V. 191, 1990, pp. 115-20.

D. Misra and S. Palmer, "Laser Ablated Thin Films of YBazCu3O,_x: the Nature
and Origin of the Particles," Physica C, 176 (1991), 43-48.

T. Nakarniya, K. Ebihara, P. John, and B. Tong, "Computer Simulation of Pulsed
Laser Ablation for YBaCuO Superconducting F ilms," Mat. Res. Soc. Symp. Proc.,
V. 191, 1990, pp. 109-114.

X. Wu, T. Venkatesan, A. Inam, X. Xi, Q. Li, W. McLean, C. Chang, D. Hwang,
R. Ramesh, L. Nazar, B. Wilkens, S. Schwartz, R. Ravi, J. Martinez, P. England,
J. Tarascon, RMuenchausen, S. Foltyn, R. Estler, R. Dye, A. Garcia, and N.
Nogan, "Pulsed Laser Deposition of High T, Superconducting Thin Films: Present
and Future," Mat. Res. Soc. Symp. Proc., V. 191, 1990, pp. 129-139._

0. Eryu, K. Murakarni, K. Masuda, K. Shihoyama, and T. Mochizuki, "Strong
Wavelength Dependence of Laser Ablation Fragments of Superconductor
YBazCu307," Jap. J. Appl. Phys. V. 31, N. 2A, Feb. 1,1992, pp. L86-8.

[51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

[62]

84

A. Cheenne, J. Perriere, F. Kerhereve, G. Hauchecome, E. Fogarassy, C. Fuchs,
"Laser Assisted Depsition of Thin BiSrCaCuO F ilms," Mat. Res. Soc. Symp. Proc.
Vol. 191, 1990, pp. 229-234.

G. Koren, A. Gupta, R. Baseman, M. Lutwyche,a nd R. Laibowitz, "Laser
Wavelength Dependent Properties of YBazCu3O7,x Thin Films Deposited by Laser
Ablation, Appl. Phys. Lett. 55, 1989,pp. 2450-2.

L. Wiedeman and H. Helvajian, "UV Tunable Laser Ablation of YBazCu3Ox,,:
Changes in the Product POpulation and Kinetic Energy Distributions as a Function
of the Laser Wavelength and Target Bulk Temperature," Mat. Res. Soc. Symp.
Proc. V. 191, 1990, pp. l99-204.

L. Wiedeman and H. Helvajian, "Threshold Level Laser Ablation of YBaZCu3O,,,,5
at 351 nm, 248 nm, and 193 nm: Ejected Product Population and Kinetic Energy
Distributions," Mat. Res. Soc. Symp. Proc. V. 191, 1990, pp. 217-222.

A. Gupta, B. Braren, K. Casey, B. Hussey, and R. Kelly, "Direct Imaging of the
Fragments Produced During Excimer Laser Ablation of YBazCu,O7_x," Appl. Phys.
Lett. 59 (l 1), Sept. 9, 1991, 1302-1304.

T. Okada, N. Shibamaru, Y. Nakayama, and M. Maeda, "Investigations of
Behavior of Particles Generated from Laser-Ablated YBa2Cu3O7,x Target Using
Laser-Induced Fluorescence," Appl. Phys. Lett. 60 (8), Feb. 24, 1992, 941-943.

H. Sakeek, T. Morrow, W. Graham, and D. Walmsley, "Optical Absorption
Spectroscopy Study of the Role of Plasma Chemistry in YBaZCu3O7 Pulsed Laser
Deposition, Appl. Phys. Lett., 59 (27), Dec. 30, 1991, pp 3631-3.

W. Weimer, "Plasma Emission from Laser Ablation of the High-Temperature
Superconductor YBa2Cu3O7," Appl. Phys. Lett. 52(25), June 20,1988,pp.2171-21 73.

T. Venkatesan, X. D. Wu, A. Inam, and J. Wachtrnan, "Observation of Two
Distinct Components During Pulsed Laser Deposition of High T, Superconducting
Films," Appl. Phys. Lett. 52 (14), April 4, 1988, pp. 1193-1195.

D. Goehegan and D. Mashbum, "Spectroscopic and Ion Probe Characterization of
the Transport Process Following Laser Ablation of Y832Cu30x," Mat. Res. Soc.
Symp. Proc. V. 191, 1990, pp.211-216.

K. Saenger, "Time-Resolved Optical Emission During Laser Ablation of Cu, CuO,
and High-T, Superconductors: BiHSrmCazCu30x and YBa,,,Cu2_7oy," J. Appl.,
Phys. 66 (9), Nov. 1, 1989,pp. 4435-4439.

J. S. Blakemore, Solid State Physics, c. 1985, pp.251-6.

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

[73]

[74] .

175]

85

J. Asmussen, "Electron Cyclotron Resonance Microwave Discharges for Etching
and Thin-Film Deposition," J. Vac. Sci. Technol., A 7 (3), May/Jun 1989.

A. Srivastava, M. Dahimene, T. Grotjohn, J. Asmussen, "Oprational Performance
of a Compact Coaxial ECR Plasma Source for MBE Applications," Wavemat Inc.
internal publication, 1991 .

V. Matijesevic, E. Garwin, and R. Hammond, "Atomic Oxygen Detection by
Silver-Coated Quartz Depositon Monitor," submitted to Appl. Phys. Lett.

D. Blank, D. Adelerhof, J. F lokstra, and H. Rogalla, "Parameter Study of In-Situ
Grown Superconducting YBaCuO Thin Films Prepared by Laser Ablation,"
Physica C, 124-126, (1989), pp. 125-126.

J. Frohlingsdorf, P. Leiderer, R. Feile, W. Zander, B. Stritzker, "Fast One Step
Prepartion of High Quality YBazCu3O.,,x Thin Films by Laser Ablation," Physica
C, 162-164, (1989), pp. 123-124.

D. Fork, K. Char, F. Bridges, S. Tahara, B. Lairson, J. Boyce, G. Connell, T.
Geballe, "YBCO Films on YSZ and A1203 by Pulsed Laser Deposition," Physica
C, 162-164,(1989), pp. 121-122.

H. U. Haberrneier and G. Mertens, Physica C, 162-164, (1989), pp. 601-2.

D. Li, K. Wang, D. Q. Li, R. Chang, J. Ketterson, "Microstructure Studies of
Epitaxial YBaQCu3O7.x Films," J. Appl. Phys. 66 (11), Dec. 1, 1989.

T. Venkatesan, E. Chase, X. Wu, A. Inam, C. Chang, and F. Shokoohi, Appl.
Phys. Lett., 53 (3), July 18, 1988, p.243.

R. Russo, R. P. Reade, J. M. McMillan, and B. L. Olsen, "Metal Buffer Layers
and Y-Ba-Cu-O Thin Films on PT and Stainless Steel Using Pulsed Laser
Depostion," J. Appl. Phys, 68 (3), Aug. 1, 1990, p 1355.

A. Carim, S. Basu, and R. Muenchausen, "Dependence of Crystalline Orientation
on Film Thickness in Laser-Ablated YBazCu307,x on LaAlO3," Appl. Phys. Lett.,
58 (8), Feb. 25 1991.

F. Yang, E. Narumi, S. Patel, and D. T. Shaw, "Weak Link Dominated Anisotropy
of Critical Current Density in Polycrystalline YBazCu307,x Thin Films," Appl.
Phys. Lett., 60 (2), Jan. 13, 1992, pp. 249-251.

M. Ohlcubo, T. Kachi, and T. Hioki, "Epitaxial YBazCu3Ox Thin Films with x=6-7
by Oxygen In-Diffusion following Laser Deposition, J. Appl. Phys. 68 (4), Aug.
15, 1990, pp. 1782-1786. ‘

1761

[77]

[781

[791

[80}

[81]

[82]

[83]

[84]

86

D. Shi, J. Krucpzak, M. Tang, N. Chen, and R. Bhadra, "Oxygen Diffusion and
Phase Transformation in YBaZCu3O7,x," J. Appl. Phys. 66 (9), Nov. 1, 1989, pp.

4325-4328.
T. B. Lindemer, F. Washbum, C. MacDougall, R. F eenstra, and O. Cavin, Physica

C, 178, 93, (1991).
R. Hammond and R. Bormann, "Correlation Between the In Situ Growth

Conditions of YBCO Thin Films and the Thermodynamic Stability Criteria,"
Physica C, 162-164,(1989) 703-4.

D. Chrisey, K. Grabowski, and M. Osofsky, "Pulsed Laser Deposition of
YBazCu307,x in an Oxygen Background and Discharge," Physica C, 162-

164(1989), pp. 129-130.
M. Itoh, H. Ishigaki, and K. Demizu, "Influence of Carbon on Critical
Temperatures of Superconducting YBaZCu3O,_x, Jap. Jour. of Appl. Phys, V.28,

N.9, Sept, 1989, pp. L1527-1530.
Y. Zhao, W. Chu, D. Christen, E. Jones, M. Davis, J. Wolfe, S. Deshmukh, and
D. Economou, "Linewidth Dependence of Critical Current Density in YBaQCu3O7

Thin-Film Microbn'dges," Appl. Phys. Lett. 59 (9), Aug. 26, 1991.

C. Chang, X. Wu, R. Ramesh, X. Xi, T. Ravi, T. Venkatesan, D. Hwang, R
Muenchausen, S. Foltyn, and N. Nogar, "Origin of Surface Roughness for C-Axis

Oriented Y-Ba-Cu-O Superconducting F ilms," Appl. Phys. Lett. 57 (17), Oct. 22,

1990, pp. 1814-6.
R. Ramesh, A. Inam, D. Hwang, T. Sands, C. Chang, D. Hart, "Surface Outgrowth

Problem in C-Axis Oriented Y-Ba-Cu-O Superconducting Thin F ilrns," Appl. Phys.

Lett. 58 (14), April 8, 1991, pp. 1557-1559.
A. Berezin, S. Pan, E. Ogawa, R. Silver, and A. De Lozanne, "Thin Films of
YBaCuO Grown In-Situ By Co-Evaporation and Plasma Oxidation, Physica C,

162-164 (1989), pp. 657-8.

[85] J. Asmussen, Private communication.

l’lIlCHI

1111111111111111111111111111111I111111

 

3008779195

 

 

 

 

\ 1|
\ , I
-.
II
II
II
.\
.5 II
V ,,
#00: II
or .
QM. -
ell-Illllll w
uIllllllIlIl-I

HICHIGQN STQTE

11311 111111111111111

779195