University of Wollongong Thesis Collections
University of Wollongong Thesis Collection
University of Wollongong
Year 
A study of the magnetoresistance effect
in Bi-2212 for the purpose of utilisation
in magnetic field sensors
Brad Winton
University of Wollongong
Winton, Brad, A study of the magnetoresistance effect in Bi-2212 for the purpose of utilisation in magnetic field sensors, M.Sc. thesis, Institute for Superconducting and Electronic
Materials, University of Wollongong, 2005. http://ro.uow.edu.au/theses/318
This paper is posted at Research Online.
http://ro.uow.edu.au/theses/318
A study of the magnetoresistance effect in
Bi-2212 for the purposes of utilisation in
magnetic field sensors
A thesis submitted in partial fulfilment of the
requirements for the award of the degree
Masters of Science - Research
From
UNIVERSITY OF WOLLONGONG
By
Brad Winton, BSc (Physics) Hons (Adv)
ISEM
2005
1
Acknowledgements
A number of people aside from my supervisors Professor Dou and Dr Mihail
Ionescu helped and contributed to the project and the writing of this thesis.
The helpful discussions with Dr Konstantin Konstantinov, Dr Josip Horvat and Mr
Greg Tillman combined with their technical expertise in the running of equipment
ensured that this project ran smoothly and on time.
2
Abstract
The magnetoresistive (MR) effect in Bi-2212 was studied through the creation of a
number of Bi-2212 composites for the purposes of utilization as the active
component in a magnetic field sensor. Bi-2212/USr2CaO6, Bi-2212/MgO, Bi2212/CeO2 and Bi-2212/Ag composites were made with varying concentrations of
impurities. The MR effects of these composites were characterized and compared to
that of pure Bi-2212 and that of the other composites.
The resistivity of the Bi-2212 composites with the exception of the Bi-2212/CeO2
and Bi-2212/Ag can be characterized by a high sensitivity to applied magnetic field
over an increasing temperature range with increasing impurities concentrations up to
a percolative limit. Bi-2212/CeO2 composites display a highly insulating behavior
indicative of a breached percolative limit while the Bi-2212/Ag composite displays a
decreased resistance as the Ag improves conductivity with the HTS.
X-ray results show the compatibility of the added impurities with the stability of the
Bi-2212 matrix in all cases except CeO2 addition during the melt texturing (MT)
process. Bi-2212/USr2CaO6 and Bi-2212/MgO showed the most potential for
cryogenic magnetic field sensing applications, and so Bi-2212 + 6.6wt% USr2CaO6
MT composite and Bi-2212 + 20wt% MgO MT composite were tested at 77K and
cycling fields of 0-1T. They showed high sensitivity to applied magnetic fields and
no hysteresis.
3
Bi-2212 + 20wt% MgO MT composite bulk proved to be the most promising
composite for the purposes of cryogenic magnetic field sensing, and so a prototype
cryogenic sensor was built with Bi-2212 + 20wt% MT composite bulk as the active
component.
4
Contents
Acknowledgements………………………………………………………………....2
Abstract…………………………………………………………………………..…..3
List of Publications resulting from this work.....................................................................8
Table of Figures......................................................................................................... 9
Table of tables .................................................................................................................17
Chapter I: Introduction ....................................................................................................18
Chapter II: Literature Review ..........................................................................................21
2.1 Superconducting magnetic field sensors for Cryogenics................................ 21
2.2 Superconducting Junctions and their use in applications ............................... 23
2.3 Types of Josephson Junction type Weak links ............................................... 28
2.3.1 SNS type weak links ............................................................................... 30
2.3.2 SIS type weak links ................................................................................ 32
2.4 Percolate systems........................................................................................... 34
2.5 Bi2Sr2Can-1CunO2(2+n)+δ (Bi-2212) .................................................................. 37
2.5.1 Transport properties of polycrystalline Bi2212....................................... 38
2.6 Compound additions in BSCCO, behaviour, compatibility............................ 41
2.6.1 Uranium-Strontium-Calcium-Oxide (USr2CaO6) Additions ................... 41
2.6.2 Magnesium Oxide (MgO2) Additions ..................................................... 42
2.6.3 Cerium Oxide (CeO2) Additions ............................................................. 43
2.6.4 Bi-2212 with Silver (Ag) Additions........................................................ 43
2.7 Possible-encapsulating materials ................................................................... 45
5
2.7.1 Polyurethane ........................................................................................... 45
2.7.2 Epoxy Resin............................................................................................ 47
Chapter III: Experimental ................................................................................................49
3.1 Bi-2212/USr2CaO6 composites ...................................................................... 51
3.2 Bi-2212/MgO composites.............................................................................. 54
3.3 Bi-2212/CeO2 composites.............................................................................. 57
3.4 Bi-2212/Ag composites ................................................................................. 59
3.5 Possible Encapsulation Package .................................................................... 61
Chapter IV: Results and Discussion ................................................................................63
4.1 USr2CaO6....................................................................................................... 63
4.1.1 Materials Characterization – XRD.......................................................... 63
4.1.2 Materials Characterization – DTA .......................................................... 66
4.1.3 Materials Characterization – SEM .......................................................... 68
4.1.4 Transport Properties – 4 probe................................................................ 72
4.2 Bi-2212/MgO Composites.........................................................................................86
4.2.1 Materials Characterization – XRD.......................................................... 86
4.2.2 Materials Characterization – DTA .......................................................... 91
4.2.3 Materials Characterization– SEM ........................................................... 92
4.2.4 Transport Properties – 4 probe................................................................ 94
4.3 Bi-2212/CeO2 Composite............................................................................. 112
4.3.1 Materials Characterization – XRD........................................................ 112
4.3.2 Materials Characterization – DTA ........................................................ 116
4.3.3 Magnetic Properties – PPMS ................................................................ 118
6
4.3.4 Transport Properties – 4 probe.............................................................. 120
4.4 Bi-2212/Ag Composites .............................................................................. 125
4.4.1 Material Characterization – XRD ......................................................... 125
4.4.2 Materials Characterization – DTA ........................................................ 128
4.4.3 Transport Properties – 4 probe.............................................................. 130
4.5 Search for a suitable encapsulation material. ............................................... 139
Chapter V: Conclusion ..................................................................................................140
Chapter VI: Sensor Fabrication .....................................................................................142
Chapter VII: Conclusion................................................................................................144
References .....................................................................................................................146
7
List of Publications resulting from this work
•
Cryogenic magnetic field sensor based on the magnetoresistive effect in
bulk bi-2212 + USr2CaO6, M. Ionescu, B. Winton, T. Silver, S. X. Dou, R.
Ramer. Applied Physics Letters, 84 (26) (2004) 5335-5338
•
Large magnetoresistive effect in bulk Bi2212 with small additions of
USr2CaO6, M. Ionescu, B. Winton, T. Silver, S. X. Dou, R. Ramer.
Journal of Applied Physics D: Applied Physics 37 (2004) 1727-1731
•
Magnetoresistive effects in Bi-2212 melt textured bulk with MgO additons,
B. Winton, M. Ionescu, T. Silver, S. X. Dou. J. Phys. D: Appl. Phys. 38
(2005) 2327-2331.
8
Table of Figures
Figure 1.1: Magnetoresistive effect seen in a granular Bi (Pb)-22235. ............... 20
Figure 1.2: Resistance Vs. Temperature plot comparing the behaviour of a
normal metal to a superconductor................................................................ 20
Figure 2.1: Schematic of a Josephson junction.................................................... 23
Figure 2.2: Schematic of randomly aligned superconducting grains separated
by non-superconducting grain boundaries. ................................................. 24
Figure 2.3: Schematic of the different conduction paths within a HTS
superconducting grain and the crystal structure of Bi-2212. ..................... 39
Figure 2.4: The Ag-CuOx system at P(O2) = 0.21 bar. ........................................ 44
Figure 2.5: A typical polyurethane compound created through the reaction
between 4,4-diisocyanatodiphenylmethane and ethylene glycol................. 47
Figure 2.6: A typical epoxy molecule made from bisphenol-A and
epichlorohidrin. ............................................................................................. 48
Figure 3.1: The temperature and pressure conditions used for sample
preparation..................................................................................................... 50
Figure 3.2: A schematic of the prototype superconducting magnetic field sensor.
......................................................................................................................... 61
Figure 4.1: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of USr2CaO6 from 0wt%-11wt%. ....................................... 64
Figure 4.2: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of USr2CaO6 from 0wt%-11wt%............. 66
9
Figure 4.3: Differential thermal analysis of the Bi-2212 powder mixtures with
USr2CaO6 concentrations from 0wt%-11wt%. The endothermal peaks are
offset against an arbitrary unit on the vertical scale. .................................. 68
Figure 4.4: Elemental mapping of the surface of a Bi-2212 + 8.8wt% USr2CaO6
composite with the electron beam ⊥ texture. ............................................... 70
Figure 4.5: An overlay of the elemental maps of Ca and U onto the image of the
surface of a Bi-2212 +8.8wt% USr2CaO6 composite with the electron beam
⊥ texture. ........................................................................................................ 71
Figure 4.6: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk. ......................................... 73
Figure 4.7: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 5.5wt% USr2CaO6 MT composite
bulk. ................................................................................................................ 74
Figure 4.8: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% USr2CaO6 MT composite
bulk. ................................................................................................................ 75
Figure 4.9: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 8.8wt% USr2CaO6 MT composite
bulk. ................................................................................................................ 76
Figure 4.10: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 11wt% USr2CaO6 MT composite
bulk. ................................................................................................................ 77
10
Figure 4.11: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 MT bulk. ................................... 79
Figure 4.12: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 5.5wt% USr2CaO6 MT bulk. 80
Figure 4.13: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 6.6wt% USr2CaO6 MT bulk. 81
Figure 4.14: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 8.8wt% USr2CaO6 MT bulk. 82
Figure 4.15: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 11wt% USr2CaO6 MT bulk. . 83
Figure 4.16: The field dependence of the relative variation in (ρr) at 77K, for
Bi-2212 + 0, 5.5wt% USr2CaO6, 6.6wt% USr2CaO6, 8.8wt% USr2CaO6 and
11wt% USr2CaO6, respectively..................................................................... 84
Figure 4.17: The change of sensor resistivity with cycled applied magnetic field
for Bi-2212 + 6.6wt% USr2CaO6 at 77K. ..................................................... 86
Figure 4.18: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of MgO from 0wt%-35wt%. Also included is the spectrum
of an untextured Bi-2212 + 15wt% MgO pellet. .......................................... 88
Figure 4.19: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of MgO from 0wt%-35wt%. .................... 90
Figure 4.20: Differential thermal analysis of the Bi-2212 powder mixtures with
MgO concentrations from 0wt%-35wt%. The endothermal peaks are
offset against an arbitrary unit on the vertical scale. .................................. 92
11
Figure 4.21: Elemental mapping of the surface of a Bi-2212 + 6.6wt% MgO
composite with the electron beam ⊥ texture. ............................................... 93
Figure 4.22: An overlay of the elemental maps of Bi, Ca and Mg onto the image
of the surface of a Bi-2212 + 6.6wt% MgO composite with the electron
beam ⊥ texture. .............................................................................................. 94
Figure 4.23: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk. ......................................... 97
Figure 4.24: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% MgO MT composite bulk.
......................................................................................................................... 98
Figure 4.25: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% MgO MT composite bulk.
......................................................................................................................... 99
Figure 4.26: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% MgO composite bulk.... 100
Figure 4.27: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 20wt% MgO MT composite bulk.
....................................................................................................................... 101
Figure 4.28: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 35wt% MgO MT composite bulk.
....................................................................................................................... 102
Figure 4.29: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 MT bulk. ................................. 104
12
Figure 4.30: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 6.6wt% MgO MT composite
bulk. .............................................................................................................. 105
Figure 4.31: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 15wt% MgO MT composite
bulk. .............................................................................................................. 106
Figure 4.32: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 15wt% MgO composite bulk.
....................................................................................................................... 107
Figure 4.33: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 20wt% MgO MT composite
bulk. .............................................................................................................. 108
Figure 4.34: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 35wt% MgO MT composite
bulk. .............................................................................................................. 109
Figure 4.35: The field dependence of the relative variation in (ρr) at 77K, for Bi2212 + 0, 6.6wt% MgO, 15wt% MgO, 20wt% MgO and 35wt% MgO MT
bulk, respectively. The ρr for untextured Bi-2212 + 15wt% MgO is also
included for comparison.............................................................................. 110
Figure 4.36: The change of sensor resistivity with cycled applied magnetic field
for Bi-2212 + 20wt% MgO MT at 77K. ..................................................... 112
Figure 4.37: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of CeO2 from 0wt%-15wt%. ............................................. 114
13
Figure 4.38: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of CeO2 from 0wt%-15wt%. .................. 116
Figure 4.39: Differential thermal analysis of the Bi-2212 powder mixtures with
MgO concentrations from 0wt%-15wt%. The endothermal peaks are
offset against an arbitrary unit on the vertical scale. ................................ 118
Figure 4.40: Magnetic Tc of Bi-2212 + 15wt% MT composite bulk................. 119
Figure 4.41: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk. ....................................... 121
Figure 4.42: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% CeO2 MT composite bulk.
Applied field ┴ Current density. ................................................................. 122
Figure 4.43: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% CeO2 MT composite bulk.
Applied field ┴ Current density. ................................................................. 123
Figure 4.44: The dependence of Log10 ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% CeO2 MT composite bulk.
....................................................................................................................... 124
Figure 4.45: The dependence of Log10 ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% CeO2 MT composite bulk.
....................................................................................................................... 125
Figure 4.46: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of Ag from 0wt%-35wt%. Also included is the spectrum of
an untextured Bi-2212 + 15wt% Ag pellet. ................................................ 127
14
Figure 4.47: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of Ag from 0wt%-15wt%. ...................... 128
Figure 4.48: Differential thermal analysis of the Bi-2212 powder mixtures with
Ag concentrations from 0wt%-15wt%. The endothermal peaks are offset
against arbitrary units on the Y-axis.......................................................... 130
Figure 4.49: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk. ....................................... 132
Figure 4.50: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% Ag MT composite bulk.
....................................................................................................................... 133
Figure 4.51: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% Ag MT composite bulk. 134
Figure 4.52: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 MT bulk. ................................. 135
Figure 4.53: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 6.6wt% Ag MT composite
bulk. .............................................................................................................. 136
Figure 4.54: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 15wt% Ag MT composite bulk.
....................................................................................................................... 137
Figure 4.55: The field dependence of the relative variation in (ρr) at 77K, for Bi2212 + 0, 6.6wt% Ag and 15wt% Ag MT composite bulk respectively. .. 138
15
Figure 4.56: The field dependence of the relative variation in (ρr) at 77K, for Bi2212, 6.6wt% Ag and 20wt% MgO and 5.5wt% USr2CaO6 MT composite
bulk respectively. ......................................................................................... 141
Figure 4.57: A schematic of the prototype superconducting magnetic field
sensor. ........................................................................................................... 143
16
Table of tables
Table 3.1: A list of the Bi-2212 + USr2CaO6 composites produced to investigate
the magnetoresistance effect of Bi-2212 with USr2CaO6 additions. ........... 52
Table 3.2: A list of the Bi-2212 + MgO composites produced to investigate the
magnetoresistance effect of Bi-2212 with MgO additions. .......................... 55
Table 3.3: A list of the Bi-2212 + CeO2 composites produced to investigate the
magnetoresistance effect of Bi-2212 with CeO2 additions........................... 58
Table 3.4: A list of the Bi-2212 + Ag composites produced to investigate the
magnetoresistance effect of Bi-2212 with Ag additions............................... 60
Table 4.1: Typical lattice parameter values deduced experimentally for the Bi2212/ USr2CaO6 composites. ......................................................................... 64
Table 4.2: Typical lattice parameter values deduced experimentally for the Bi2212/ MgO composites................................................................................... 89
Table 4.3: Typical lattice parameter values deduced experimentally for the Bi2212 + CeO2 composites. ............................................................................. 114
Table 4.5: Typical lattice parameter values deduced experimentally for the Bi2212 + Ag composites................................................................................... 127
17
Chapter I: Introduction
Since the earliest days of magnetic field sensing when navigators used the earth’s
magnetic poles to explore the oceans, the industry of magnetic field sensing has
expanded enormously. With the benefit of not having to be in physical contact to
deliver results, magnetic field sensors have been adapted to measure small magnetic
fields in such diverse applications as the detection of the presence of DNA or
antibodies1 and magnetic ink character recognition (MICR).
While there are numerous sensors available for the measurement of both large and
small fields at room temperature for a vast range of applications, there remain
limited technologies for the measurement of magnetic fields at cryogenic
temperatures, particularly around the boiling point of liquid nitrogen (77K)2.
Currently the magnetic field sensors commonly used at such low temperatures are
those based on the Hall effect or the magnetoresistive effect in semiconductors3.
The Hall effect based sensors and the magnetoresistive-based sensors are both
expensive, however, leading to the exploration of other more commercially viable
technologies, including colossal magnetoresistive (CMR) and spintronic based
magnetic field sensors.
CMR materials display much larger magnetic field dependences than other
magnetoresistive materials such as giant magnetoresistance materials and anisotropic
18
magnetoresistance devices. With as much as one million percent change in
resistance based on a semiconductor-metal transition at the Curie temperature (TC)
being reported, the magnetoresistance measured in CMR is by far superior to that of
other magnetoresistors. Due to the reduced sensitivity and a strong non-linearity of
the MR effect at cryogenic temperatures (≅77K) in these materials, however, the
possible applications of CMR-based cryogenic magnetic sensors may be limited.
Another emerging technology is the use of spintronics for the purposes of magnetic
field sensing. While still in the development stage, spintronic magnetic field sensors
have the potential to be extremely sensitive while working in a wide temperature
range by using the spin orbit coupling effect in a quantum wire4. Working from 4.2K
to room temperature a spintronic magnetic field sensor works on the basis of
alternating spin relaxation times. A magnetic field drastically reduces the spin
relaxation time; this can be used to determine an unknown magnetic field.
Nevertheless, spintronic sensors are still years away from being commercially
available, so there is room for an alternative technology in the meantime.
There has been some interest within the field of superconductivity in producing such
an alternative. It should be noted that many polycrystalline high–temperature
superconductors (HTS) show a magnetoresistive (MR) effect, but this effect is
usually small and within a narrow temperature range close to their critical
temperature (Tc)5 as is shown in Fig. 1.1, resulting in the potential of HTS for
19
magnetic field sensing being frustrated. The critical temperature of a superconductor
is the temperature at which the dc resistivity drops to zero (Fig. 1.2).
Please see print copy for Figure 1.1
Figure 1.1: Magnetoresistive effect seen in a granular Bi (Pb)-22235.
Figure 1.2: Resistance Vs. Temperature plot comparing the behaviour of a
normal metal to a superconductor.
20
Chapter II: Literature Review
2.1 Superconducting magnetic field sensors for Cryogenics
While researching superconductivity, Brian David Josephson studied the properties
of a junction between two superconductors. Following up on earlier work by Leo
Esaki and Ivar Giaever6, he demonstrated that there was an electron flow between
two superconductors through an insulating layer, something he had theorised in
19627 (Fig. 2.1).
This has consequences for this study since a weak link is created when a barrier of
non-superconducting material separates two superconducting grains within a
polycrystalline HTS material. If superconducting tunnelling occurs between the
grains and through the non-superconducting barrier the weak links inhibit the
behaviour of the so-called Josephson junction. It should be noted that weak link or
Josephson junction behaviour does not exist if the non-superconducting barrier
between superconducting grains inhibits tunnelling because of its dimensions.
The magnetoresistive behaviour seen in polycrystalline HTS materials is the product
of their granular nature, with the native grain boundaries forming weak link barriers
to electric currents passing between the superconducting grains. When a
supercurrent passes through weak links by tunneling through a non-superconducting
barrier they can be described as Josephson junctions as seen above. It is for this
reason that polycrystalline HTS are sensitive to applied magnetic fields. A
21
polycrystalline HTS can be expected to become more sensitive to magnetic fields
with increasing concentrations of Josephson junctions as more of the supercurrent is
reliant on passing through the weak links within the polycrystalline material.
In previous studies it has been shown that doping Y123 with BaPb1-xSnxO3 in an
attempt to create a network of Josephson junctions by introducing a large number of
artificial grain boundaries leads to a significant increase in the MR effect at low
magnetic fields when compared to Y1238, 9. Due to strong intragrain pinning
attributed to Y123, however, a hysteretic behaviour was reported for the measured
MR upon cycling of the applied magnetic field.
Other non-superconducting additions to HTS materials have shown that the increase
in the magnetorestive effect is not limited to Y123 nor always hindered by a
hysteretic behaviour. An increase in the MR effect in Bi-2212 with the addition of
USr2CaO addition was seen with no such hysteretic effect, making it a viable
alternative for use in magnetoresistive magnetic field sensors10 and other potential
applications. This procedure was also used for a HTS of another family with the
addition of CuO and BaPbO3 producing similar non – hysteretic MR effects 11.
22
Figure 2.1: Schematic of a Josephson junction.
2.2 Superconducting Junctions and their use in applications
HTS bulk materials are not, in general, a homogeneous continuum but rather a
network of linked superconducting grains12. The reason for this polycrystalline
structure is the particular mechanism for their crystal growth. All of the material that
cannot fit into the lattice structure of the growing native superconducting grains is
pushed forward into the growth front with the consequence that in the end all of the
remnants of secondary phases and impurities are concentrated at the boundaries
between the superconducting grains (Fig. 2.2).
Since the discovery of HTS13,
14
materials, studies have found that the transport
properties of polycrystalline HTS materials are affected by these native grain
boundaries forming weak links between the superconducting grain colonies15,
16
.
Distinct from the strong links which are responsible for carrying supercurrents at
higher fields, weak links are easily decoupled by magnetic fields and are responsible
for the initial rapid decrease of critical current density (Jc) with applied magnetic
23
field (H). While weak links can be created from the addition of a nonsuperconducting phase and secondary phases between the superconducting grains as
mentioned previously, they could also be created by twist boundaries and large angle
tilt boundaries 17, 18.
Careful control of the growth process and in particular of the composition of the
starting material can significantly reduce the amount of non-superconducting
material and secondary phases present between the grain boundaries at the end of
processing. By minimizing the amount of non-superconducting material and
secondary phases present between the grain boundaries weak links are minimized
and the superconducting properties within these polycrystalline HTS structures are
maximized with the creation of more strong links.
Figure 2.2: Schematic of randomly aligned superconducting grains separated
by non-superconducting grain boundaries.
24
It is not always desirable, however, to minimize the number of weak links in a
superconducting sample by removing all of the non-superconducting or secondary
phases. Superconducting devices have a wealth of applications as electronic devices
through the manipulation of the superconductor microstructure.
When the non-superconducting and secondary phases fail to produce optimal results
in producing a useful network of Josephson junctions it may be necessary to add
other foreign materials into the superconducting compound or even replace the
impurities with a predetermined addition. When a foreign phase is deliberately
inserted into the superconducting precursor or ready-made powder before
processing, a superconducting composite can result. This is where two separate
phases exist independently within the final superconducting matrix. In HTS
composites the non-superconducting materials usually added are normal metals (N),
insulating materials (I) or degenerate semiconductors (Sm).
Manipulation of the microstructure for optimal results can go further than just
adding additional non-superconducting phases to introduce artificial grain
boundaries. By optimising processing conditions and the initial stoichiometry of the
superconductor non-superconducting and secondary phases can be controlled and
suppressed, leaving the added foreign material to be pushed into the growth front. In
the process of growing a superconducting lattice with a known foreign material
concentrated at and around the boundaries between the superconducting grains
without other unknown impurities affecting the weak coupling between
25
superconducting colonies, the random network of Josephson junctions can be
controlled19, 20.
It cannot be said that all foreign material added to the superconducting phase within
the bulk material creates Josephson junction weak links. If too much of the material
accumulates within the grain boundaries the non-superconducting phase might start
to inhibit current flow and prevent superconducting tunnelling. When this happens
the barrier can no longer be described as a Josephson junction, and the barrier ceases
to contribute to the magnetoresistive effect of the entire sample.
Since HTS materials are not isotropic and have short coherence lengths the
superconducting properties of HTS materials are susceptible to structural and
chemical changes on atomic length scales. As a result, transport across the
artificially introduced barriers of non-superconducting material is highly dependent
on microstructural imperfections in the barrier and at its interface. This raises
questions about the reproducibility of results because of the difficulty in keeping the
composition of the barrier and its interface identical.
Despite this, HTS composites have a large role in applications as artificial networks
of weak link Josephson junctions. Josephson junction devices such as SQUIDs and
other electronic devices based on HTS technology are in an advanced and constant
state of development around the world. Of particular interest though is the proximity
effect, which is central to the physics of the coupling of superconductivity across
26
non-superconducting boundaries that make the Josephson junctions used in HTS
electronics possible. The proximity effect involves the reciprocal influence of
neighbouring superconducting and non-superconducting materials across an
interface.
In conventional Low–Tc (LTS) superconductors the proximity effect is relatively
well understood for interfaces with normal metals21 due to a combination of the
BCS22 theory and long coherence lengths. Typical of LTS materials, these long
coherence lengths average out and temper interface effects, allowing for the use of
simple phenomenological boundary conditions.
The very short coherence lengths which characterize HTS materials make them
much more susceptible to the influence of neighbouring materials and internal
defects. This makes the use of simple phenomenological models problematic,
leading to a reliance on a poorly developed microscopic theory and thus a limited
understanding of the proximity effect in HTS. This is why when creating a HTS
composite it is useful to have minimized the secondary phases and other unknown
non-superconducting phases present within the grain boundaries.
A strong doping dependence of the cuprate superconductors makes them particularly
sensitive to charge transfers at interfaces where there is a tendency to form npn- like
junctions23, thus introducing a further complexity. Associated with each type of HTS
27
composite material, are random Josephson junction networks of SNS, SSmS or SIS
lattices, with unique phenomena driving their characteristic behaviour.
2.3 Types of Josephson Junction type Weak links
Electronic structure and charge transport across these artificially created weak links
are determined by the type and strength of symmetry breaking leading to different
behaviours between the types of lattices. As well as spontaneously broken gauge
symmetry, responsible for diagonal long-range order, phase coherence and
superconductivity
24, 25
, there are two other types of broken translational symmetry
The first is due to the complex pair potential ∆(r), associated with off-diagonal longrange order, which varies spatially, so “broken off diagonal translational symmetry”
may be spoken of. Secondly, special variations of the diagonal scalar and vector
potentials U and A result in “broken diagonal translational symmetry.”
In an ideal SNS junction, a micro-bridge where the lattice structures and the Fermi
energies are the same in the S layers and the N region, spatial variations of normal
state properties can be neglected. Only off-diagonal translational symmetry is
broken in this case.
For the SIS junctions with exponential dampening of the wave amplitudes in the
insulating layer there is a strong breaking of translational symmetry. In this case
28
superconducting coupling is so weak that the Josephson effects can be calculated to
the second order.
Finally in SSmS junctions both diagonal and off-diagonal symmetry breaking exists
because of the interface potentials and very different lattice structures and carrier
concentrations between the S and the Sm layers.
Because of the unique phenomena driving the behaviour of Josephson junction
networks produced by artificially created weak links, HTS based composites hold
fundamental interest. Transport properties of HTS + N 26, 27, 28, 29, 30, 31, HTS + Sm 32
and HTS + I
33, 34, 35, 36
composites have been widely investigated because of their
potential for possible practical applications as previously mentioned, but also as an
object for the study of intergrain coupling,
37, 38, 39, 40, 41
. Despite the depth of
investigations into artificially created weak links, high-Tc cuprate Josephson
junctions remain mostly SNS networks since SIS junctions do not show a strong
Josephson effect 42.
There are some fundamental properties that need to be considered when using HTS
intrinsic junctions such as SNS and SIS in Josephson junction networks for
cryogenic application purposes. Of particular interest to this project when using Bi2212 composites for use as the core in a magnetic field sensor is the non-hysteretic
R(H,T) characteristic behaviour in the random array of Josephson junctions; this
leads to controllable and reproducible parameters. Another characteristic of
29
importance for any future cryogenic sensor is high stability at room temperature and
under thermal cycling, and impact resistance.
2.3.1 SNS type weak links
As mentioned earlier, a SNS weak link is created when a normal metal layer lies
between two superconducting crystallites creating an artificial grain boundary. This
can be achieved through sub-micron fabrication or, as in the case of this study,
through the production of a HTS composite bulk. A HTS composite bulk has
artificial weak links that are created though the existence of a non-superconducting
separate phase at the grain boundaries.
When a supercurrent tunnels through this artificial grain boundary the SNS weak
link behaves like a Josephson junction. Although early studies of artificially created
grain boundaries focused mainly on the SIS type junctions43, 44, with the discovery of
a new scattering mechanism at the SN boundary known as Andreev reflection the
importance of SNS junctions rose45. As mentioned earlier, if the boundary potential
is too large for tunnelling to exist then there will be no Josephson junction. A fine
balance must then be struck when determining how thick a boundary is optimal
between superconducting grains.
The Josephson effect in SNS weak links is responsible for the well-known
appearance of macroscopic phase coherence, the mechanism by which the phase
30
coherent electrons are transported from one superconductor to another through a
non-superconducting phase, in this case when superconducting crystallites are
weakly coupled to each other. In these structures, the dc Josephson effect arises due
to coherent charge transport in the normal region, the mechanism of which is
Andreev reflection, which may be seen as the transport of Cooper pairs from one
superconducting electrode to another 46, 47, 48.
Andreev reflection is where an electron in the normal region of the SNS junction
incident on the SN interface is reflected back as a hole, and visa versa in reverse.
This can be interpreted as the condensation of the incident electron together with
another electron, corresponding to the reflecting hole, into a Cooper pair or in the
case of an incoming hole, as the disassociation of a Cooper pair.
This behaviour of SNS type weak links initiated intensive study with first generation
calculations derived by Kulik 49, 50, Ishii 51, Bardeen and Johnson 52, and Svidzinsky
et al.
53
. As a result of this new type of weak link a new wealth of physics and
applications of Josephson junctions was developed. Gunsenheimer et al.54 attempted
to establish a unified macroscopic understanding of SNS junctions taking into
account Cooper pair tunneling, the proximity effect and Andreev reflection.
If a SNS junction were to be used successfully in any sort of future application, it
would be necessary to know how such a junction behaves under a magnetic field
since any individual effect would be amplified within a network of such junctions.
31
It is well known that magnetic field in general suppresses supercurrent and its effects
on SNS junctions provide an interesting study. Suppression of supercurrent via
magnetic field can arise from two completely different mechanisms. Firstly, it can
result from coupling of the magnetic field to superconductivity via the vector
potential, however, this effect can be suppressed in geometries where the area’s
component perpendicular to the magnetic field is sufficiently small.
The second mechanism is due to the Zeeman energy. Since pairing is a singlet in swave superconductors, and the Zeeman field is a pair breaking perturbation, the
Cooper pair amplitude decays in space faster in the presence of the field. For SNS
junctions, the supercurrent thus decreases with field due to the reduction of coupling
between the two superconductors. It is interesting to note though, that under a
suitable non-equilibrium distribution of quasiparticles the suppressed supercurrent
can be recovered. A suitable applied voltage can partly recover the supercurrent
suppressed by magnetic field in a SNS junction.
2.3.2 SIS type weak links
HTS + insulator composites form a similar network of random Josephson Junctions
to HTS + normal metals. Like the case with HTS + N composites the insulating
material forms artificial grain boundaries between the superconducting crystallites
which can behave like Josephson junctions if tunneling occurs, and many of the
32
same ideas about how these junctions work apply equally to SIS junctions. There are
however some fundamental differences, and it is important to understand them.
Their behaviour under magnetic fields and the Josephson effect differs from SNS
junctions and needs to be discussed.
The relative lack of Josephson effect mentioned earlier in connection with SIS weak
links is closely related to the dependence of the maximum DC Josephson current on
the tunneling resistance (Rn) found in low Tc superconductors55,
56, 57, 58
. Above
several ohms the Josephson current decreases much faster than 1/Rn. Unlike single
particle tunneling, the tunneling of Cooper pairs can lead to a finite supercurrent
(Josephson current) flowing across the SIS junction in the absence of a bias voltage.
It is advantageous to deal with the Josephson effect in HTS with insulating barriers
of high tunneling resistance from the Cooper pair wave function approach since the
Cooper pair tunneling can be understood more easily.
It can be shown that the Josephson coupling energy is determined by the overlap of
the Cooper pair wave functions of the two superconductors divided by the thin
insulating layer. The critical current is then calculated from the coupling energy in
the usual way 59. From this model it is found that the Cooper pair tunneling is
strongly limited by the size of the Cooper pair, which is reduced during the
tunneling. Since the impurity potential scattering also reduces the Cooper pair size,
ordinary impurities also limit the supercurrent.
33
All this acts to reduce the Josephson current to much less than expected for high Tc
SIS junctions. For thicker and higher insulating barriers the Josephson current falls
off much faster than 1/Rn in good agreement with experimental findings
60
. HTS
have tightly bound Cooper pairs with small size, which leads to difficulty tunneling
through an insulating barrier. This difficulty can only lead to the small critical
currents seen in SIS Josephson junctions.
Another striking differences between SIS and SNS type junctions is the way in
which they behave under magnetic fields. It has already been shown that
supercurrent flowing through a SNS network exhibits the Zeeman effect, while the
critical current density through SIS Josephson structures shows a Fraunhofer
diffraction pattern at a constant magnetic field. This is made possible by the
relatively low Josephson current and low self-field. It is a consequence of the wave
like nature of the Cooper pairs and the phase coherence through the junction. The
extreme sensitivity of the supercurrent is an important feature in many important
applications
61
. Like the SNS weak links, however, stronger fields gradually
decrease the coupling between the two superconducting crystallites, suppressing
supercurrent through the junction and network in general and limiting the
supercurrent to any existing strong links within the bulk material.
2.4 Percolate systems
Various HTS composites have been made through the addition of foreign nonsuperconducting phases at different concentrations in order to experimentally
34
compare different types of Josephson junctions26,
27, 28, 29, 30, 31
. Technologically,
HTS-metal composites offer attractive features, which make them useful in a wide
area of applications, while as previously mentioned HTS-insulator composites offer
attractive opportunities for the study of intergrain coupling.
Experimentally however, a composite of HTS and a second alien nonsuperconducting phase has proven difficult to achieve without severely
compromising the superconductive properties. At the high temperatures required to
prepare HTS ceramics most foreign materials tend to react with the ceramic
compounds, degrading their superconductivity by promoting the growth of
secondary and non-superconducting phases at the expense of the HTS phase.
There are, however, phases that have a limited chemical effect in the HTS materials
even when processed at high temperatures and pressures. For those that don’t
adversely affect the parent-superconducting matrix by degrading its phase identity,
percolative behaviours in both the normal state and the superconducting state have
been observed. USr2CaO6, MgO, CeO2 and Ag have proved ideal for the creation of
Bi-2212 composites.
The percolation concept was first used to describe superconductors in the mid
seventies62 when Davidson and Tinkham analysed resistivity data of composite
Nb3Sn/Cu wires. Percolation theory has also been used in the analyses of a number
of superconducting/non-superconducting composites prepared by powder mixing
35
and sintering63. The basic assumption in percolation theory is that the state of an
individual grain, grain boundary or phase boundary is controlled by random
fluctuations independently of its neighbors.
When the superconductor – normal transition is induced by a transport current this
basic assumption doesn’t apply64, 65; here the Kirchhoff rules must be considered,
and the probability of individual grain boundaries becoming non-superconducting is
no longer independent. Percolation theory cannot be applied easily in this case
because correlation effects are expected. Despite the difficulty in applying the
theory, it is a useful tool in analysing HTS composite systems.
It is widely understood that the transport properties of HTS composite percolating
systems depend on the volume fraction (ρ) of the superconducting constituent. The
critical concentration (ρc) required for superconductivity (percolation threshold)
depends on the shape of the particles and the dimensionality of the composite. For ρ
≥ ρc the superconducting grains bond together into a connected superconducting
lattice, which spans the entire system. Conversely when ρ ≤ ρc the metal host forms a
connected metallic network spanning the system. For a HTS composite the
percolation threshold has been experimentally determined to be approximately 0.266
volume percent.
Thus by adjusting the volume fraction of the superconductor between ρc and 1 - ρc,
the electrical and mechanical properties of the HTS composite percolating system
36
can tuned between a metal like state and a ceramic like state while still maintaining
overall superconductivity within the composite. A detailed study of current transfer
through polycrystalline superconductor by percolation theory together with its
limitations can be found in Zeimetz et al.67
2.5 Bi2Sr2Can-1CunO2(2+n)+δ (Bi-2212)
There are a few HTS materials that have a good potential for electronic applications.
Among them are two from the BSCCO family, Bi-2212 and Bi-2223 (first
generation materials). One is from the YBCO family, Y-123, and there remains the
possibility for Hg based and Tl based superconductors to be used in certain specific
applications (third generation of applied superconductor materials)68.
Bi-2212 from the first generation BSCCO family remains the most promising
because of some advantageous materials properties including:
1. A low concentration of weak links;
2. High density is easily obtained from liquid phase;
3. Extremely stable with regards to temperature and chemical composition;
4. Chemical stability of ready made phase;
5. Absence of toxic elements; and
6. Critical temperature above that of the boiling point for liquid nitrogen;
37
Recently Marinkovic et al.
69, 70
demonstrated the viability of a partial melt process
whereby the Bi-2212 is melted at temperatures slightly greater than the peritectic
melting temperature, followed by slow cooling to produce bulk with good
superconducting properties. Previous studies of melt texturing71 allow for a
thoroughly researched and easily scaled process for producing bulk Bi-2212 in
larger scale commercial operation should the viability of Bi-2212 composites be
proven for applications.
It is for the above reasons that Bi-2212 was chosen over other superconductive
materials for this study. If results in the laboratory warranted the production of Bi2212 bulk for use in superconducting applications it would be easily translated into a
commercial setting.
2.5.1 Transport properties of polycrystalline Bi2212
Superconductivity in HTS materials, specifically Bi-2212, is believed to originate in
the physics of the CuO2 layers where the mobile charges are located72,
73
(fig. 4).
Interlayer coupling between these CuO2 layers within a given (CuO2/Ca/)n-1CuO2
stack is much weaker than the intralayer coupling within the CuO2 layers. Despite
this, interlayer coupling is still much stronger than the coupling between
(CuO2/Ca/)n-1CuO2 stacks which can be described as Josephson coupling. This
quasi-2-dimensional nature of superconductivity in HTS materials leads to a
38
prominent anisotropy of the superconducting properties with much higher
supercurrents along the CuO2 planes than in the perpendicular direction (Fig. 2.3).
Figure 2.3: Schematic of the different conduction paths within a HTS
superconducting grain and the crystal structure of Bi-2212.
Bi2Sr2Can-1CunO2(2+n)+δ has a varying oxygen concentration 74, 75 which is manifested
in the Bi-O layers 76, 77, 78, 79. It is well known that the oxygen content of the cuprate
is related to the physical properties and structure of the superconducting material. A
direct relationship can be established between the amount of intercalated oxygen, the
cell parameter c and the optimising of Tc through the removal of oxygen by proper
annealing or further oxidation 80. The crystal lattice parameters for Bi-2212 are
a=b=5.410nm and c=30.84nm. However, a small change in oxygen content can have
a dramatic effect on not only the Tc but also the magnitude of the lattice parameter c.
39
In solid-state physics, a particle's effective mass is the mass it seems to carry in the
transport in a crystal. Under most conditions carriers, both electrons and holes
respond to electric and magnetic fields in a crystal almost as if they were free
particles in a vacuum, but with a different mass. This mass is usually stated in units
of the ordinary mass of an electron me.
Effective mass is defined by analogy with Newton's second law. Using quantum
mechanics it can be shown that for an electron in an external electric field E:
where a is acceleration, h is Planck's constant, k is the wave number, ε(k) is the
energy as a function of k, this is often called the dispersion relation. From the
external electric field alone, the electron would experience a force of qE, where q is
the charge. Hence under the model that only the external electric field acts, effective
mass m* becomes:
Bi-2212, like the other compounds in the Bi2Sr2Can-1CunO2(2+n)+δ system shows a
high level of anisotropy with the conducting CuO2 planes stacked between oxide
40
layers. Anisotropy can be expressed in terms of a ratio between effective masses
with γ = (mc/m(a, b))^1/2, for Bi-2212 γ has been reported between 55 and 200 81, 82, 83.
2.6 Compound additions in BSCCO, behaviour, compatibility
USr2CaO6, MgO, CeO2 and Ag have proved ideal for the creation of Bi-2212
composites, and their effects on Bi-2212 phase and microstructure have been studied
to various degrees in the past.
2.6.1 Uranium-Strontium-Calcium-Oxide (USr2CaO6) Additions
Unlike Ag, the effect of USr2CaO6 addition to Bi-2212 hasn’t been exhaustively
studied. Originally used as an alternative to UO2, U3O8 and UCa2O5 as a vehicle for
adding elemental Uranium to Bi-2223 for the purposes of the U/n method84, the
addition of the compound UCa2SrO6 into Bi-2212 has been largely ignored. Like Bi2223, adding Uranium oxide into Bi-2212 has been found to be detrimental. At
concentrations greater than 0.4wt% U, the oxide was found to have a poisoning
effect on the superconducting properties of the material by adversely affecting the
microstructure.
It has been found that the compound USr2CaO6 minimizes the harmful effects of the
interaction of Uranium on the Bi-2212 microstructure to such an extent that the
superconducting properties tolerate increased additions of up to 3wt% U85. While
this is important in the area of superconductivity, for the sake of this project the
41
UCa2SrO6 ability to maintain its identity through the high temperatures and
pressures involved in processing Bi-2212 makes it more relevant. This ability to
remain separate from the parent matrix makes it a useful non-magnetic insulator to
add to Bi-2212 in an attempt to increase the MR effect by increasing the number of
grain boundaries in a created SIS network of Josephson type weak links.
2.6.2 Magnesium Oxide (MgO2) Additions
In 1992 Soylu et al86 reported that Bi-2212 did not react with MgO intensifying the
study of MgO addition to Bi-2212 87,88, 89. S-L. Huang et al reported a significant
reduction in both the volume and population of Sr-Ca-Cu-O secondary phases with
the addition MgO and suggested that MgO addition may have a role at changing the
phase evolution of Bi-2212 during partial melting and solidification. This behavior
was put down to the MgO sites playing the roles of particular centers with locally
changed characteristics meaning that melting process starts at many centers
simultaneously90 rather than at a single site.
Although its been reported that MgO addition degrades grain alignment, this can be
expected from any foreign addition into the Bi-2212 matrix including UCa2SrO6,
CeO2 and Ag. Like Ag, UCa2SrO6 and CeO2 though, it’s chemical stability during
the high temperature and pressures of processing Bi-2212 make this an ideal
candidate for addition into the Bi-2212 bulk with the purposes of creating artificial
Josephson junctions weak links.
42
2.6.3 Cerium Oxide (CeO2) Additions
CeO2 is mainly used as a buffer layer between substrates such as sapphire and
biaxially textured Ni-alloys and HTS thin films because of its chemical stability
under high pressure and temperatures, the fact that it doesn’t absorb liquid phase Bi2212, and because it has a lattice mismatch of less than 1%91 at room temperature. It
is for these reasons that CeO2 is particularly useful as a Bi-2212 thin film substrate,
and they make it a likely candidate for addition into the Bi-2212 bulk.
Lattice mismatch isn’t nearly as important in polycrystalline bulk as it is in thin
films, but nevertheless the other advantages of CeO2 allow for near optimal
processing conditions for Bi-2212, leaving room for a random network of Josephson
junction type weak links between Bi-2212 superconducting crystallites and highly
insulating CeO2 ceramic powder.
2.6.4 Bi-2212 with Silver (Ag) Additions
The effect of non-magnetic silver (Ag) on Bi-2212 bulk is well known and has been
exhaustively studied. In the processing of Bi-2212, Ag has long been recognized as a
unique non-poisoning noble metal
92
which reduces the partial melting temperature,
enhances the formation of superconducting phases, increases the mechanical
ductility and decreases the room temperature resistivity93. More recently it was
reported that Ag dissolves in the liquid at the partial melt stage94, giving rise to
further investigations on the effects of Ag on the microstructure of Bi-2212 since the
43
method of partial melting is primarily used to obtain Bi-2212 bulk with decent
superconducting properties.
The presence of Ag within close proximity to Bi-2212 promotes the formation of
copper rich and silver rich dendritic phases in Bi-2212 bulk during cooling from
partial melt conditions. Both Ag and Cu rich dendrites arise from near the silver
particles leading to a possible inhomogeneity in microstructure further from the
silver presence. The lowering of the partial melting temperature is thought to be able
to suppress the generation of non-superconducting phases within the Bi-2212/Ag
composite, leading to a greater control over microstructure. This control can be seen
in the reported Ag-Cu2O-CuO phase diagram, which also shows Ag rich and Cu rich
compositions (Fig. 2.4).
Please see print copy for Figure 2.4
Figure 2.4: The Ag-CuOx system at P(O2) = 0.21 bar95.
44
It is possible to optimize processing conditions such that the interaction between the
Bi-2212 and Ag within the composite will be limited to the benefits of Ag use,
leaving the non-superconducting metal to coat the superconducting grains or fill the
grain boundaries depending on the amount of Ag added to the polycrystalline HTS.
This should ensure that a network of SNS junctions will be grown homogeneously
throughout the material.
2.7 Possible-encapsulating materials
As mentioned previously, in the process of creating a superconducting magnetic
field sensor it is important that it be both stable under thermal cycling and impact
resistant for ease of use. To provide some impact resistance and facilitate ease of use
of the superconducting magnetic field sensor it is important to encapsulate the active
component of HTS composite in a buffer material. Both polyurethane and epoxy
resin were suggested as candidates for this purpose.
2.7.1 Polyurethane
Polyurethane elastomers are rubber-like materials that can be created with a wide
variety of properties and molded into almost any shape. Formed by reacting a dialcohol with a diisocyanate or a polymeric isocyanate (Fig 2.5) in the presence of
suitable catalysts and additives their physical properties can be manipulated
45
depending on their intended use by varying the constituents. A variety of
diisocyanates and a wide range of di-alcohols can be used to produce polyurethane
elastomers with a varying resistance to:
1. Abrasion,
2. Impact and shock,
3. Temperature,
4. Cuts and tears,
5. Oil and solvents,
6. Aging,
7. Mold, mildew and fungus, and
8. Most types of chemicals.
Because the physical properties of polyurethane can be manipulated by varying the
diisocyanates and di-alcohols used, a broad spectrum of materials can be produced
to meet the needs of specific applications.
Polyurethanes exist in a variety of forms including flexible foams, rigid foams,
chemical-resistant coatings, specialty adhesives and sealants, and elastomers and are
used in a variety of ways including insulation, cushioning and even sealants for
construction and applications where their high strength, moisture resistance and
durability are required.
46
Figure 2.5: A typical polyurethane compound created through the reaction
between 4,4-diisocyanatodiphenylmethane and ethylene glycol.
2.7.2 Epoxy Resin
The word epoxy was coined from a Greek word meaning ‘over’ or ‘between’ and
oxygen because of its unique oxygen between compound structures. An illustration
of this can be seen in Figure 2.6 where the value of n, the repeating unit, is less than
1 for the lowest molecular weight liquid resins. Increasing n not only increases the
molecular weight of the resin but also increases its viscosity until it finally solidifies.
47
Figure 2.6: A typical epoxy molecule made from bisphenol-A and
epichlorohidrin.
It has been said that no plastic since the invention of celluloid has created such
curiosity, opportunity or confusions as epoxy resins. This is because epoxies unlike
other plastics offer an infinite array of chemical, physical and electrical properties
along with a wide choice of hardening agents, curing cycles and methods of
compounding, handling, storing and application.
Where no epoxies existed to suit an application, new epoxies have been developed.
The electronics and silk screening and printing industries have benefited enormously
while new epoxies have been specially designed for applications in optics.
Paradoxically, the use of exotic materials like specially designed epoxies often prove
to reduce the costs of production due to the reduction in production times, while
eliminating scrap and often enabling a process that would not have been possible
without such a material.
While the array of chemical and physical properties of epoxies is impressive by
filling the epoxy with metallic and non-metallic powders the properties can be subtly
changed further expanding its possible uses.
48
Chapter III: Experimental
High temperature superconductors need to be textured in order to show large critical
current densities. Intergranular junctions inducing weak links limit the
superconducting properties of HTS96, so texturing aligns grains according to their ab
planes, improving intergranular junctions. For this reason the hot pressing method
has been developed both to sinter a pellet of ceramic HTS and to improve its
texturing. This has ramifications when the magnetoresistance effect relies heavily on
links between grains.
The hot press is a furnace built around a mechanical press that applies a uniaxial
pressure during sintering, dramatically increasing the density of the processed pellet.
An electronic controller attached to the furnace controls the temperature conditions
ensuring that the furnace remains at an optimal temperature for melt texturing. Melt
texturing is a process that involves partially melting the material while applying
pressure so that a small amount of liquid phase co-exists with solid phase. To ensure
that no chemical reaction takes place between the HTS pellet and the alumina disks
during melt texturing, a silver foil is placed between the two alumina disks while
Al2O3 based slurry between the alumina disks and the silver sheets buffers the
effects of different coefficients of expansion that may produce cracking of the
sample upon cooling.
49
Figure 3.1: The temperature and pressure conditions used for sample
preparation.
A series of Bi-2212 composites were produced with this process for the purpose of
exploring the magnetoresistance effect through the addition of various foreign
materials of different concentrations. After dry mixing a small quantity of
stoichiometric Bi-2212 powder, produced by spray drying, with small quantities of
the chemically compatible USr2CaO6, MgO, CeO2 and Ag, the mixtures were then
pelletized and melt textured in air under a uniaxial force of 900N.
Usually when adding one unknown material to another, an exhaustive optimization
study is needed to obtain optimal sintering conditions. Certainly the addition of
foreign materials to Bi-2212 needed investigation so that the effect of different
concentrations of additions could be partially understood and a new melting point
determined. By using a DTA analysis of the melting profile for each of the mixtures,
50
however, this exhaustive study, which would have consumed the majority of the
time allotted for this project, was bypassed.
While it could be said that the sintering processes obtained from the differential
thermal analysis were not the optimal outcome, it was a fair approximation that
allowed for the focus of this work to shift from materials fabrication to the
magnetoresistive effect of Bi-2212 composites. This MT process produced disc
shaped Bi-2212 composite pellets with a typical density of 92%-94% theoretical
with the ab plane of the Bi-2212 unit cell perpendicular to the applied force.
3.1 Bi-2212/USr2CaO6 composites
Four Bi-2212/USr2CaO6 composites with varying concentrations of USr2CaO6
addition were created for the purposes of studying the MR effect of Bi-2212 with
USr2CaO6 additions. Three grams of Bi-2212 + USr2CaO6 was thoroughly dry
mixed to insure maximum homogeneity of the varying concentrations of USr2CaO6
within the ready made stoichiometric Bi-2212. The powder mixtures with 5wt%,
6wt%, 8wt% and 11wt% USr2CaO6 were then pelletised and melt textured at
temperatures predetermined by DTA, producing disc-shaped Bi-2212/USr2CaO6
composite pellets with a typical density of 92%-94% theoretical.
51
Table 3.1: A list of the Bi-2212 + USr2CaO6 composites produced to investigate
the magnetoresistance effect of Bi-2212 with USr2CaO6 additions.
Sintering
Composite
Dimensions
Temperature
Material
lxwxh (mm)
(oC)
Bi-2212
7x2x2
843
5.44x1.88x1.57
843
8.81x1.87x1.50
843
5.63x2.54x2.31
843
6x2.37x1.67
843
Bi-2212 + 5wt%
CeO2 Textured
Bi-2212 + 6wt%
CeO2 Textured
Bi-2212 + 8wt%
CeO2 Textured
Bi-2212 + 11wt%
CeO2 Textured
The Bi-2212/USr2CaO6 MT composite pellets were characterized with XRD
techniques to ensure that the phase purity of the composite constituents was
conserved after processing, and rocking scans were done on peaks (008) and (0010)
to investigate the effects of USr2CaO6 addition to Bi-2212 on the out of plane
alignment and the composite crystallinity. Lattice parameters a and c were
experimentally determined for the Bi-2212 structure to investigate the presence and
amount of any interaction between foreign phases and the host material.
52
It is of interest in the study of the MR effect in Bi-2212 composites to investigate the
homogeneity of the addition within the final composite. To this effect a fragment of
Bi-2212 + 9wt% USr2CaO6 composite was embedded in an epoxy mould. The
surface perpendicular to the applied force was polished to 0.25 micron. SEM and
EDX were then done on the carbon coated surface of the polished Bi-2212 + 8wt%
USr2CaO6 chip.
A rectangular sample was dry cut from the MT composite pellets for transport
measurements using the standard 4-probe method. The field dependence of the
resistivity of the samples was then measured between 5K and room temperature in
fields up to 9T. This was done using a Quantum Design Physical Properties
Measurement system (PPMS) with the applied magnetic field perpendicular to the
ab plane.
Keeping in mind the Bi-2212 + USr2CaO6 composite’s potential role in magnetic
field sensing, the variation of the resistance of textured Bi-2212 + 6wt% USr2CaO6
with cycling applied field was measured. This was done at 77K with I=5mA in low
magnetic fields as a function of increasing and decreasing applied fields between 0T
– 1T as measured by the PPMS. A significant hysteresis effect would drastically
limit the usefulness of the composite in the field of magnetic sensing at cryogenic
temperatures, so this measurement was vital in ensuring that the Bi-2212 +
USr2CaO6 composite was a viable alternative in the field of magnetic sensing.
53
3.2 Bi-2212/MgO composites
Five Bi-2212 + MgO composites with varying concentrations of MgO addition were
created for the purposes of studying the MR effect of Bi-2212 with MgO additions.
Three grams of Bi-2212 + MgO was thoroughly dry mixed to insure maximum
homogeneity of the varying concentrations of MgO within the ready-made
stoichiometric Bi-2212 powder. Four of the powder mixtures with 6wt%, 15wt%,
20wt% and 35wt% MgO were then pelletised and melt textured, producing disc
shaped Bi-2212 + MgO composite pellets with a typical density of 92%-94%
theoretical. A fifth powder mixture of 15wt% MgO was pelletised and sintered in a
tube furnace rather than the hot press so that the effects of melt texturing and
increased density on the MR of Bi-2212 + MgO composites could be isolated.
54
Table 3.2: A list of the Bi-2212 + MgO composites produced to investigate the
magnetoresistance effect of Bi-2212 with MgO additions.
Sintering
Dimensions
Temperature
Composite Material
lxwxh (mm)
(oC)
Bi-2212
7x2x2
843
7.27x2.65x2.22
844
5.8x2.25x0.4
839
5.8x2.25x0.4
874
8.27x3.55x2.7
835
13x1.3x4.7
825
Bi-2212 + 6wt% MgO
Textured
Bi-2212 + 15wt% MgO
Textured
Bi-2212 + 15wt% MgO
Untextured
Bi-2212 + 20wt% MgO
Textured
Bi-2212 + 35wt% MgO
Textured
The Bi-2212 + MgO MT composite pellets were characterized with XRD techniques
to ensure that the phase purity of the composite constituents was conserved after
processing, and rocking scans were done on peaks (008) and (00 10) to investigate
the effects of MgO addition on the Bi-2212 out of plane alignment and composite
crystallinity. Lattice parameters a and c were experimentally determined for the Bi-
55
2212 structure to investigate the presence and amount of any interactions with
foreign phases.
It is of interest in the study of the MR effect in Bi-2212 composites to investigate the
homogeneity of the addition within the final composite. To this effect a fragment of
Bi-2212 + 6wt% MgO composite was embedded in an epoxy mould. The surface
perpendicular to the applied force during processing was polished to 0.25 micron
with diamond paste. SEM and EDX were done on the carbon coated surface of the
polished Bi-2212 + 6wt% MgO fragment.
A rectangular sample was dry cut from the MT Bi-2212 + MgO composite pellets
for transport measurements using the standard 4-probe method. The field
dependence of the resistivity of the samples was then measured between 5K and
room temperature in fields up to 9T. This was done using a Quantum Design
Physical Properties Measurement system (PPMS) with the applied magnetic field
perpendicular to the ab plane.
Keeping in mind the Bi-2212 + MgO composite’s potential role in magnetic field
sensing, the variation of the resistance of textured Bi-2212 + 20wt% MgO with
cycling applied field was measured. This was done at 77K with I=5mA in low
magnetic fields as a function of increasing and decreasing applied fields between 0T
– 1T as measured by the PPMS. A significant hysteresis effect would drastically
limit the usefulness of the composite in the field of magnetic sensing at cryogenic
56
temperatures, so this measurement was vital in ensuring that the Bi-2212 + MgO
composite was a viable alternative in the field of magnetic sensing.
3.3 Bi-2212/CeO2 composites
Another non-toxic, non-magnetic insulator, which was reported to be chemically
compatible with Bi-2212, was CeO2. So in a further investigation of the MR effect in
Bi-2212, two Bi-2212/CeO2 composites with varying concentrations of CeO2
addition were created for the purposes of studying the MR effect of Bi-2212 with
varying additions of CeO2. Three grams of Bi-2212 + CeO2 was thoroughly dry
mixed to insure maximum homogeneity of the varying concentrations of CeO2
within the ready-made stoichiometric Bi-2212. The powder mixtures with 6wt% and
15wt% CeO2 were then pelletised and melt textured, producing disc-shaped Bi-2212
+ CeO2 composite pellets with a typical density of 92%-94% theoretical.
57
Table 3.3: A list of the Bi-2212 + CeO2 composites produced to investigate the
magnetoresistance effect of Bi-2212 with CeO2 additions.
Sintering
Composite
Dimensions
Temperature
Material
lxwxh (mm)
(oC)
Bi-2212
7x2x2
843
9x1.3x1.5
872
8.73x2.5x2.876
872
Bi-2212 + 6wt%
CeO2 Textured
Bi-2212 + 15wt%
CeO2 Textured
The Bi-2212 + CeO2 MT composite pellets were characterized with XRD techniques
to ensure that the phase purity of the composite constituents was conserved after
processing, and rocking scans were done on peaks (008) and (00 10) to investigate
the effect of CeO2 addition on the Bi-2212 out of plane alignment and composite
crystallinity. Lattice parameters a and c were experimentally determined for the Bi2212 structure to investigate the presence and amount of any interactions with
foreign phases.
A rectangular sample was dry cut from the MT Bi-2212 + CeO2 composite pellets
for transport measurements using the standard 4 probe method. The field
dependence of the resistivity for the samples was then measured between 5K and
room temperature in fields up to 9T in a similar way to the Bi-2212 + MgO MT
pellets. This was done using a Quantum Design Physical Properties Measurement
58
system (PPMS) with the applied magnetic field perpendicular to the ab plane. A
direct comparison could then be drawn between the MR effect generated by CeO2
addition and the addition of the other materials studied in this work.
3.4 Bi-2212/Ag composites
With the majority of this study focusing on Bi-2212 + insulator composites it was
interesting to compare how a Bi-2212 + metal would behave in the area of
magnetoresistivity. For this reason two Bi-2212/Ag composites with varying
concentrations of Ag addition were created for the purposes of studying the MR
effect on Bi-2212 of additions of Ag. Three grams of Bi-2212 + Ag was thoroughly
dry mixed to insure maximum homogeneity of the varying concentrations of
powdered Ag within the ready-made stoichiometric Bi-2212. The powder mixtures
with 6wt% and 15wt% Ag were then pelletised and melt textured, producing discshaped Bi-2212 + Ag composite pellets with a typical density of 92%-94%
theoretical.
59
Table 3.4: A list of the Bi-2212 + Ag composites produced to investigate the
magnetoresistance effect of Bi-2212 with Ag additions.
Sintering
Composite
Dimensions
Temperature
Material
lxwxh (mm)
(oC)
Bi-2212
7x2x2
843
7.35x2.25x2.26
819
8.6x2.29x1.77
819
Bi-2212 + 6wt%
Ag Textured
Bi-2212 + 15wt%
Ag Textured
The Bi-2212 + Ag MT composite pellets were characterized with XRD techniques to
ensure that the phase purity of the composite constituents was conserved after
processing, and rocking scans were done on peaks (008) and (00 10) to investigate
the effect of Ag addition on Bi-2212 out of plane alignment and composite
crystallinity. Lattice parameters a and c were experimentally determined for the Bi2212 structure to investigate the presence and amount of any interaction with foreign
phases.
A rectangular sample was dry cut from the MT Bi-2212 + Ag composite pellets for
transport measurements using the standard 4-probe method. The field dependence of
the resistivity for the samples was then measured between 5K and room temperature
in fields up to 9T. This was done using a Quantum Design Physical Properties
60
Measurement system (PPMS) with the applied magnetic field perpendicular to the
ab plane.
3.5 Possible Encapsulation Package
As mentioned earlier it is important for the superconducting magnetic field sensor to
be both easy to use and resistant to thermal shocks and pressure variations. For this
reason the HTS composite was embedded in an encapsulating mould, creating a selfcontained unit as seen below in Figure 3.2.
Figure 3.2: A schematic of the prototype superconducting magnetic field sensor.
Two possible candidate materials where chosen for this, polyurethane and aluminum
filled epoxy resin. The silver back plate and base in direct contact with the active
component of the magnetic sensor ensured that whatever the encapsulating material
61
was made of the HTS composite would maintain thermal equilibrium with its
environment.
Blocks of polyurethane and epoxy resin were made up and repeatedly thermally
cycled between liquid nitrogen temperature and room temperature to check their
resistance
to
thermal
shock.
62
Chapter IV: Results and Discussion
4.1 USr2CaO6
4.1.1 Materials Characterization – XRD
The X-ray analysis on the Bi-2212 MT disks with various USr2CaO6 concentrations
is presented in Figure 4.1. By comparing the spectra of the MT pellets it can be seen
that each is well aligned in the (00l) direction with the added USr2CaO6 phase
identified as a separate phase against the host matrix. There is no drifting of the (00l)
Bi-2212 peaks, and the experimentally determined lattice parameters (Table 4.1)
only show evidence of a decrease in intercalated oxygen with increasing
concentration of USr2CaO6. There is no evidence that there is any pollution of the
Bi-2212 lattice from the diffusion of USr2CaO6 addition. This is a reliable
confirmation
that
USr2CaO6 retains
its
phase
independence
within
the
superconducting composite, despite the high temperature and pressures exerted
during processing.
63
I (cps)
I (cps)
800
400
0
I (cps)
300
200
100
0
I (cps)
800
400
0
I (cps)
008 0010
0012
6000
4000
2000
0
2000 0
1500
1000
500
0
10
0
10
Bi-2212
* USr2CaO6
006
0
10
20
30
40
50
60
70
Bi-2212 + 5.5wt%USr2CaO6
0
10
20
30
40
*
0
10
20
40
*
50
40
*
60
40
70
Bi-2212 + 8.8wt%USr2CaO6
50
60
70
Bi-2212 + 11wt%USr2CaO6
*
30
70
Bi-2212 + 6.6wt%USr2CaO6
*
30
20
60
*
30
20
50
50
60
70
Angle (2θ)
Figure 4.1: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of USr2CaO6 from 0wt%-11wt%.
Table 4.1: Typical lattice parameter values deduced experimentally for the Bi2212/ USr2CaO6 composites.
wt% USr2CaO6
0
5.5
6.6
8.8
11.0
c (Ẳ)
30.77900
30.859
30.857
30.849
30.84
a (Ẳ)
5.41000
5.428
5.41
5.40
5.42
64
Looking at the FWHM in degrees (Fig. 4.2) of the (008) and (00 10) peaks gives an
insight into the Bi-2212/USr2CaO6 composite out of plane alignment and
crystallinity, which can then be compared to that of other concentrations of
USr2CaO6 addition. It is immediately clear from these curves that initially, the
addition of USr2CaO6 to the composite decreases out of plane alignment and the
average size of the Bi-2212 grains. This is what was expected; the uranium
compound at these low concentrations seems to find its way into the grain
boundaries and introduce new artificial boundaries decreasing the average size of the
Bi-2212 crystallites.
Between 6-8 wt% USr2CaO6 additions however, the rocking curve peaks and starts
to decrease, indicating an improving out of plane alignment and an enlargement of
the average size of the Bi-2212 crystallites. This seems to indicate that the effect of
the USr2CaO6 addition in introducing artificial grain boundaries has hit a maximum.
With increasing concentrations the USr2CaO6 uranium compound, which had
previously been located only within grain boundaries, starts to form agglomerations
as grain boundaries start to meet and merge within the composite.
This behavior decreases the number of grain boundaries within the composite
material leading to the average size of Bi-2212 grain boundaries increasing as more
and more of the added material agglomerates forming grains of foreign material. It is
expected that these large agglomerations of USr2CaO6 crystallites will have a
detrimental effect on the Bi-2212 phase and microstructure.
65
10
9
8
FWHM (theta - s)
7
6
008
5
0010
4
3
2
1
0
0
2
4
6
8
10
12
Concentration of USrCaO additions (wt%)
Figure 4.2: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of USr2CaO6 from 0wt%-11wt%.
4.1.2 Materials Characterization – DTA
Figure 4.3 shows the DTA results for the powder mixtures of ready-made Bi-2212
with USr2CaO6 additions in concentrations from 0wt%-11wt%. The endothermal
peaks represent the melting point while the broadening and shifting of the melting
profile demonstrates the influence that changing USr2CaO6 concentrations has on the
materials characteristics of Bi-2212.
66
The broadening of the endothermal peak with increased USr2CaO6 content allows
for a greater region of partial melting. This indicates that the USr2CaO6 is behaving
as a flux allowing for a higher sintering temperature to be used97. Sintering Bi-2212
at higher temperatures is widely understood to generate better outcomes due to
peritectic decomposition by minimizing the creation of non-superconducting phases
within the matrix, however complete melting will destroy the Bi-2212 phase so a
permanent compromise is required.
With the only change in the melting profile of the Bi-2212 + USr2CaO6 powder
mixes represented by the small broadening of the endothermal peak, the XRD results
are vindicated. Even with melting there remains no interaction between the two
materials. This is the best possible outcome when choosing potential materials to
add to Bi-2212 for magnetoresistance purposes and vindicates the decision to use
this compound in this project.
67
35
30
Bi-2212
Bi-2212 + 5.5 wt% USr2CaO6
Bi-2212 + 6.6 wt% USr2CaO6
Bi-2212 + 8.8 wt% USr2CaO6
Bi-2212 + 11 wt% USr2CaO6
Heat Flow (arb units)
25
20
15
10
5
0
-5
-10
-15
700
720
740
760
780
800
820
840
860
880
900
920
940
Temperature (C)
Figure 4.3: Differential thermal analysis of the Bi-2212 powder mixtures with
USr2CaO6 concentrations from 0wt%-11wt%. The endothermal peaks are
offset against an arbitrary unit on the vertical scale.
4.1.3 Materials Characterization – SEM
It is of interest to this study to be able to have a look at the orientation of the grains
in a composite and the placement of elements throughout with elemental mapping.
Figure 4.4 is an elemental map of approximately thirty microns square, showing the
general placement of a number of grains and the concentration of the various
elements contained within a Bi-2212 + 8.8 wt% USr2CaO6 composite. It is widely
68
assumed from the XRD results that material is well aligned in the (00l) direction
with the added USr2CaO6 sitting between Bi-2212 grains.
At this concentration of USr2CaO6 addition, however, the rocking curves have
started to show a decrease in FWHM indicating an increase in average Bi-2212
crystallite size. The first thing immediately obvious about Figure 4.4 is that the Bi2212/ USr2CaO6 composite lacks the well-textured look that was expected. If it were
not for earlier XRD spectra it would be tempting to say that there was no texturing
within the bulk. The agglomerations of USr2CaO6 which coexist with the Bi-2212
phase at these concentrations of addition appears to have a negative effect on grain
alignment.
69
Figure 4.4: Elemental mapping of the surface of a Bi-2212 + 8.8wt% USr2CaO6
composite with the electron beam ⊥ texture.
Calcium and Uranium have been overlaid on the image in Figure 4.5 to give a
clearer impression of what is happening within the composite. There is a clear
demonstration of calcium and uranium rich regions in the image with some calcium
rich regions overlapping uranium rich areas with strontium largely homogeneously
dispersed throughout the sample. This is indicative of regions with an increased
concentration of uranium compound.
70
At 8.8wt% USr2CaO6 addition the USr2CaO6 is clearly beginning to agglomerate
into grains of impurity phase separate from artificial grain boundaries and Bi-2212
grains. While this is occurring there also seems to be a large segregation of Ca. This
Bi deficient phase is unlikely to be superconducting and is no doubt a consequence
of such large quantities of addition with the agglomerations of the USr2CaO6 phase
promoting this segregating behavior.
Figure 4.5: An overlay of the elemental maps of Ca and U onto the image of the
surface of a Bi-2212 +8.8wt% USr2CaO6 composite with the electron beam ⊥
texture.
71
4.1.4 Transport Properties – 4 probe
The dependencies of resistivity on temperature and applied fields of 0T – 9T for the
MT bulk pellets of Bi-2212 and Bi-2212 + USr2CaO6 composites are displayed in
Figures 16-20. It can be seen that for the pure Bi-2212 MT pellet the temperature
and field dependence of the resistivity follow the expected trend as the temperature
falls from 100K, with a decrease demonstrating metallic behavior followed by a
drop at the critical temperature of approx. 93K. The temperature of the zero value of
resistivity, (ρ≤10-6 Ω cm) decreases with increasing applied field and the transition
becomes broadened in high fields.
With increasing uranium compound up to the limit of 6wt%-8wt% (Figure 4.2), a
broadening of the transition can be seen in the R(T, H) corresponding to the
increasing number of grain boundaries and impurities within the superconductor.
Past the maximum in FWHM when the USr2CaO6 reaches the maximum number of
grain boundaries, agglomerations begin to form in the material (Figure 4.5) leading
to the sharpening of the transitions with further increases in uranium compound
concentration as the average size of Bi-2212 crystallites once again starts to increase
with decreasing number of grain boundaries.
72
-3
1.0x10
-4
9.0x10
0T
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
-4
8.0x10
-4
Resistivity (Ω cm)
7.0x10
-4
6.0x10
-4
5.0x10
-4
4.0x10
-4
3.0x10
-4
2.0x10
-4
1.0x10
0.0
-4
-1.0x10
0
20
40
60
80
100
Temperature (K)
Figure 4.6: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk.
73
Resistivity (Ω cm)
1.8x10
-3
1.6x10
-3
1.4x10
-3
1.2x10
-3
1.0x10
-3
8.0x10
-4
6.0x10
-4
4.0x10
-4
2.0x10
-4
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.7: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 5.5wt% USr2CaO6 MT composite bulk.
74
-2
7.0x10
-2
6.0x10
0T
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
Resistivity (Ω cm)
-2
5.0x10
-2
4.0x10
-2
3.0x10
-2
2.0x10
-2
1.0x10
0.0
-2
-1.0x10
0
20
40
60
80
100
Temperature (K)
Figure 4.8: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% USr2CaO6 MT composite bulk.
75
-4
5.0x10
-4
4.5x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-4
4.0x10
-4
Resistivity (Ω cm)
3.5x10
-4
3.0x10
-4
2.5x10
-4
2.0x10
-4
1.5x10
-4
1.0x10
-5
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.9: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 8.8wt% USr2CaO6 MT composite bulk.
76
-4
5.5x10
-4
5.0x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-4
4.5x10
-4
Resistivity (Ω cm)
4.0x10
-4
3.5x10
-4
3.0x10
-4
2.5x10
-4
2.0x10
-4
1.5x10
-4
1.0x10
-5
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.10: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 11wt% USr2CaO6 MT composite bulk.
The broadening of the resistive transition for polycrystalline HTS in applied dc
fields has been explained previously10 with excess resistivity (ρe) being suggested as
a better way of comparing the MR effect in Bi-2212 composites. Excess resistivity at
a particular temperature (T) due only to the presence of the applied field (H) is
defined as:
ρe = ρ(T, H) - ρ(T, H=0)
(4-1)
77
Excess resistivity has been calculated using equation (4-1) for Bi-2212 and Bi-2212
+ USr2CaO6 addition with the results shown in Figures 4.11 – 4.15. A comparison
between the four Bi-2212 + USr2CaO6 composites and pure Bi-2212 reveals that the
excess resistivity or the size of the MR effect is two orders of magnitude greater in
the case of 6.6wt% USr2CaO6 compared to pure Bi-2212.
As expected the magnetoresistance of the Bi-2212/USr2CaO6 composite increases
with an increased incidence of grain boundaries. Normally the supercurrent within a
bulk superconductor will pass through the path of least resistance and distance. As
the concentration of grain boundaries increases, the ability of the supercurrent to
avoid the introduced barriers decreases and the current is forced through the network
of SIS junctions.
This behavior is responsible for the broadening of the excess resistivity peaks since
the supercurrent will be forced through the USr2CaO6 filled grain boundaries even at
lower temperatures as concentration of uranium compound is increased. Once the
USr2CaO6 starts to form agglomerations however the behavior of the resistivity is
reversed. Grain boundaries will become less common, leading to the number of
current paths that avoid the uranium compound increasing, particularly at low
temperatures. This can be seen in the re-sharpening of the excess resistivity peaks
and the reduction in the difference between applied fields.
78
It should be noted that while the USr2CaO6 grains still form a SIS junction with
surrounding superconducting grains, no tunneling occurs as a result of its size.
Without superconducting tunneling occurring, there is an evitable decrease in the
Resistivity (Ω cm)
magnetoresistive effect.
9.0x10
-4
8.0x10
-4
7.0x10
-4
6.0x10
-4
5.0x10
-4
4.0x10
-4
3.0x10
-4
2.0x10
-4
1.0x10
-4
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.11: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 MT bulk.
79
-3
1.2x10
-3
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
Resistivity (Ω cm)
1.0x10
-4
8.0x10
-4
6.0x10
-4
4.0x10
-4
2.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.12: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 5.5wt% USr2CaO6 MT bulk.
80
-2
Excess resistivity (ρe) [Ω cm]
6.0x10
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
-2
5.0x10
-2
4.0x10
-2
3.0x10
-2
2.0x10
-2
1.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.13: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 6.6wt% USr2CaO6 MT bulk.
81
-4
3.0x10
-4
2.5x10
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-4
Resistivity (Ω cm)
2.0x10
-4
1.5x10
-4
1.0x10
-5
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.14: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 8.8wt% USr2CaO6 MT bulk.
82
-4
4.0x10
-4
3.5x10
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-4
Resistivity (Ω cm)
3.0x10
-4
2.5x10
-4
2.0x10
-4
1.5x10
-4
1.0x10
-5
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.15: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 11wt% USr2CaO6 MT bulk.
For a potential magnetic sensing device the relative variation of resistivity (ρr) is an
important characteristic, which is defined as:
ρr = |ρ(H=0)/ρ(H)| x 100
(4-2)
where ρ(H) is the resistivity in applied field H and ρ(H=0) is the resistivity in the
absence of the applied dc field. Expressed as a percentage, the relative variation of
83
resistivity at 77K for Bi-2212, Bi-2212 + 5.5wt% USr2CaO6, 6.6wt% USr2CaO6,
8.8wt% USr2CaO6 and 11wt% USr2CaO6 is presented in Figure 4.16.
Relative Resistance [ρr] (%)
70
60
50
40
Bi-2212
5.5wt% USr2CaO6
6.6wt% USr2CaO6
8.8wt% USr2CaO6
11wt% USr2CaO6
30
20
10
0
0
2
4
6
8
10
H (77K)
Figure 4.16: The field dependence of the relative variation in (ρr) at 77K, for
Bi-2212 + 0, 5.5wt% USr2CaO6, 6.6wt% USr2CaO6, 8.8wt% USr2CaO6 and
11wt% USr2CaO6, respectively.
Here we can see more clearly what is happening within the composite material as the
USr2CaO6 concentration is increased through the maximum shown in Figure 4.2. As
the number of grain boundaries increases the supercurrent is forced through a
84
growing network of SIS junctions corresponding to its increasing sensitivity to
applied magnetic field. As the number of grain boundaries reaches the maximum
number within the composite, the magnetoresistance also reaches a maximum
because of the high sensitivity and ease with which an applied magnetic field can
decouple a weak link which was formerly passing supercurrent and change the
resistance of the material as the supercurrent is forced through another current path
within the composite.
This means that it would be simple enough once calibrated, to use this material’s
known resistance to establish an unknown magnetic field. To demonstrate the
concept of the Bi-2212 + 6.6wt% USr2CaO6 composite as a more viable material for
a superconducting magnetic field sensor, a prototype was constructed. The
dimensions of the active part were 8.00mm x 2.00mm with a thickness of 0.75mm
and four contacts attached using a low resistance silver paste. With the applied
magnetic field perpendicular to the ab plane of the sensor the variation in resistivity
was measured by PPMS while the applied magnetic field was cycled at 77K. The
lack of hysteretic behavior in Figure 4.17 should be noted and considered when
designing a superconducting magnetic field sensor.
85
-2
6.0x10
-2
Sensor Resistivity [Ω cm]
5.5x10
-2
5.0x10
-2
4.5x10
T=77K
I=10mA
-2
4.0x10
Field increase
Field decrease
-2
3.5x10
-2
3.0x10
-2
2.5x10
0
200
400
600
800
1000
Hall probe field [mT]
Figure 4.17: The change of sensor resistivity with cycled applied magnetic field
for Bi-2212 + 6.6wt% USr2CaO6 at 77K.
4.2 Bi-2212/MgO Composites
4.2.1 Materials Characterization – XRD
The X-ray analysis on the Bi-2212 MT disks with various MgO concentrations is
presented beside the X-ray spectra for the Bi-2212 + 15wt% MgO untextured pellet
in Figure 4.18. With the exception of the untextured pellet, from the spectra of the
MT pellets it can be seen that each is well aligned in the (00l) direction with the
added MgO phase identified as a separate phase against the host matrix. While there
86
is some limited texturing in the untextured pellet as a result of the partial melt
processing it is not significant nor in the (00l) direction.
Like the Bi-2212/USr2CaO6 composites there is no visible drifting of the (00l) Bi2212 peaks in the these x-ray spectra even though the experimentally determined
lattice parameters (Table 4.2) show evidence of some limited acceptance of Mg into
the Bi-2212 lattice with increasing addition of MgO concentration within the Bi2212/MgO composite. The variation in c seems to represent experimental error
rather than a fluctuation in intercalated oxygen within the Bi-2212 lattice. There also
appears to be an increase in secondary phases present in both the Bi-2212 + 20wt%
MgO and Bi-2212 + 35wt% MgO MT samples. Intergrowth of secondary phases is a
common problem to BSCCO phases and indicates a need to optimize the melt
texturing process for these concentrations of MgO additions.
87
I (cps) I (cps) I (cps) I (cps)
I (cps) I (cps)
008 0010 0012
6000
4000
2000
0
1200 0
800
400
0
10
0
10
2000
1000
0
006
0
10
20
30
40
50
60
70
*
300 0
200
100
0
10
20
30
40
Bi-2212 + 6.6wt%MgO
50
60
10
20
30
40
20
30
40
50
60
10
20
30
30
40
50
*
40
70
Bi-2212 + 15wt%MgO
*
60
*
20
Untextured Bi-2212 + 15wt%MgO
*
*
0
70
*
0
2000
1000
0
400
200
0
Bi-2212
* MgO
70
Bi-2212 + 20wt%MgO
*
50
60
70
*
50
60
Bi-2212 + 35wt%MgO
70
Angle (2θ)
Figure 4.18: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of MgO from 0wt%-35wt%. Also included is the spectrum of an
untextured Bi-2212 + 15wt% MgO pellet.
88
Table 4.2: Typical lattice parameter values deduced experimentally for the Bi2212/ MgO composites.
wt% USrCaO
0
6.6
15.0
20.0
35.0
c (Ẳ)
30.77900
30.88
30.852
30.874
30.854
a (Ẳ)
5.41000
5.408
5.412
5.416
5.422
Looking at the FWHM in degrees (Fig. 4.19) of the (008) and (00 10) peaks gives an
insight into the Bi-2212/MgO composite out of plane alignment and crystallinity,
which can then be compared to that of other concentrations of MgO addition within
the composite. It is immediately clear from these curves that like the previously
studied composite, initially the addition of MgO decreases out of plane alignment
and the average size of the Bi-2212 grains within the Bi-2212/MgO composite. This
is what was expected; the MgO at these low concentrations seems to find its way
into the grain boundaries and introduce new artificial boundaries decreasing the
average size of the Bi-2212 crystallites.
Between 17-20 wt% MgO additions however, the rocking curves peak and start to
decrease indicating an improvement in the out of plane alignment and an
enlargement of the average size of the Bi-2212 crystallites. This seems to indicate
that the effect of the MgO addition in introducing artificial grain boundaries has hit a
maximum. With increasing concentrations the MgO, which had previously only been
89
located within grain boundaries, starts to form agglomerations as grain boundaries
start to meet and merge within the composite.
This behavior decreases the number of grain boundaries within the composite
material leading to the average size of Bi-2212 grain boundaries increasing as more
and more of the added material agglomerates, forming grains of foreign material
within the composite.
14
12
FWHM (theta - s)
10
8
008
0010
6
4
2
0
0
5
10
15
20
25
30
35
40
Concentration of MgO addition (wt%)
Figure 4.19: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of MgO from 0wt%-35wt%.
90
4.2.2 Materials Characterization – DTA
Figure 4.20 shows the DTA results for the powder mixtures of ready made Bi-2212
and MgO additions of concentrations from 0wt%-35wt%. The endothermal peaks
represent a melting point with the broadening and shifting of the DTA melting curve
demonstrating the influence that changing MgO concentrations has on the material
characteristics of Bi-2212.
The broadening of the endothermal peak with increased MgO content allows for a
greater region of partial melting. This indicates that the MgO is behaving as a flux,
allowing for a higher sintering temperature to be used98. Sintering Bi-2212 at higher
temperatures is widely understood to generate better outcomes, however, complete
melting will destroy the Bi-2212 phase so a permanent compromise is required.
The fluctuations in the melting points of the Bi-2212 + MgO powder mixes may be
explained by in-homogeneously mixed powders. Although they were all prepared in
the same manner, a trend in either direction rather than an oscillating melting point
would have been expected had they all been homogeneously mixed. Dry mixing is
not the ideal method of homogeneously dispersing material throughout a mix and the
ceramic quality of MgO made the MgO addition increasingly hard to disperse
throughout the Bi-2212 powder. An ultrasonic stirrer would have been ideal and the
use of a non-reactive solution for wet mixing, which does not dissolve or poison
MgO or Bi-2212, is something to be investigated in the future since homogeneously
dispersed impurities would yield more reliable results.
91
Bi-2212
Bi-2212 + 6.6wt% MgO
Bi-2212 + 15wt% MgO
Bi-2212 + 20wt% MgO
Bi-2212 + 35wt% MgO
Heat flow (arb units)
30
20
10
0
-10
700
720
740
760
780
800
820
840
860
880
900
920
940
Temperatre (C)
Figure 4.20: Differential thermal analysis of the Bi-2212 powder mixtures with
MgO concentrations from 0wt%-35wt%. The endothermal peaks are offset
against an arbitrary unit on the vertical scale.
4.2.3 Materials Characterization– SEM
It is of interest in this study to look at the orientation of grains and the placement of
elements within the Bi-2212/MgO composite. Figure 4.21 is an elemental map of
approximately eight microns squared showing the general placement of a number of
grains and the concentration of the various elements contained within a Bi-2212 +
6.6wt% MgO composite. In agreement with the previous x-ray spectra for Bi2212/MgO composites, the bulk is lightly textured. Apart from a Cu rich spot in the
92
center of the image the elemental mapping shows a pure phase Bi-2212 material
with MgO concentrated in the holes between the grains as would be expected for
6.6wt% MgO addition from previous results and modeling.
Figure 4.21: Elemental mapping of the surface of a Bi-2212 + 6.6wt% MgO
composite with the electron beam ⊥ texture.
Bismuth, Calcium and Magnesium have been overlaid on the image in Figure 4.22
to give a clearer impression of what is happening within the MT bulk. There is a
clear demonstration of a copper rich region central to the image with magnesium
93
concentrated between the grains.
Bismuth is evenly distributed throughout the
grains, being poor in the holes and the Cu rich region. The other elements show a
similar distribution to Bismuth within the image. This supports the x-ray spectra and
rocking curves in showing that at this concentration MgO sits between the Bi-2212
grains.
Figure 4.22: An overlay of the elemental maps of Bi, Ca and Mg onto the image
of the surface of a Bi-2212 + 6.6wt% MgO composite with the electron beam ⊥
texture.
4.2.4 Transport Properties – 4 probe
The dependencies of resistivity on temperature and applied fields of 0T – 9T for the
MT bulk pellets of Bi-2212 and Bi-2212 + MgO composites are displayed alongside
94
the dependency of resistance on temperature and applied field for untextured Bi2212 + 15wt% MgO composite bulk in Figures 33-38. It can be seen that for the
pure Bi-2212 MT pellet the temperature and field dependence of the resistivity
follow the expected trend as the temperature falls from 100K, with a decrease
demonstrating metallic behavior followed by a drop at the critical temperature of
approx. 93K. The temperature of the zero value of resistivity, (ρ≤10-6 Ω cm)
decreases with increasing applied field, and the transition becomes broadened in
high fields.
The addition of MgO up to 6.6wt% changes the room temperature resistivity
significantly but does not alter its temperature dependence. As the addition of MgO
is increased to 15wt%, 20wt% and 35wt% the temperature dependence of the
resistivity becomes increasingly semiconductor-like. Although the onset of
superconductivity for the Bi-2212 + MgO composites is the same as for the pure Bi2212 pellets, the width of the normal-superconducting transition broadens with
increasing MgO additions. With increasing applied magnetic field this transition is
seen to broaden even further.
With the exception of the concentration (wt%) of the foreign addition the behavior
of the Bi-2212/MgO composite is remarkably similar to the behavior of Bi2212/USr2CaO6 composite. As the concentration of the MgO added to the composite
is increased the number of grain boundaries increases and the average size of Bi2212 crystallites decreases, as demonstrated by the increasing slope of the rocking
95
curves (Fig. 4.19). At 17wt%-20wt% the number of grain boundaries is at a
maximum within the sample, so further increases in MgO concentration within the
composite leads to a merging of grain boundaries and the formation of unique MgO
grains within the sample. The peak in the rocking curves (Fig. 4.19) is a
demonstration of this behavior.
This adequately explains the behavior of the MT pellets, however, when a Bi-2212 +
15wt% MgO composite was created without the use of melt texturing the R(T, H)
behavior is significantly different (Fig. 4.26). The superconducting transition is far
sharper than with the MT pellets, confirming that the melt texturing process is
significant in the demonstrated magnetoresistive effects. The explanation for this lies
in the way in which melt texturing affects where the foreign additions are placed
within the composite.
When melt texturing takes place the Bi-2212 matrix is encouraged to grow in the
(00l) plane with the ab plane aligned. As the growth front pushes the added material
in front of it into the grain boundaries, the MgO concentrations within the composite
is much less likely to meet and form agglomerations until much higher
concentrations. When there is no melt texturing involved the grains are free to grow
in whatever direction is most favorable; this means there is an increased likelihood
that grain boundaries will meet and form agglomerations of MgO earlier, allowing
the supercurrent to bypass the SIS junctions in favor of other paths with less
resistance.
96
Intergrowths of secondary BSCCO phases enhances this effect as seen in the Bi2212 + 20wt% MgO and Bi-2212 + 35wt% MgO MT R H, T) results seen bellow.
-3
1.0x10
-4
9.0x10
0T
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
-4
8.0x10
-4
Resistivity (Ω cm)
7.0x10
-4
6.0x10
-4
5.0x10
-4
4.0x10
-4
3.0x10
-4
2.0x10
-4
1.0x10
0.0
-4
-1.0x10
0
20
40
60
80
100
Temperature (K)
Figure 4.23: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk.
97
Figure 4.24: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% MgO MT composite bulk.
98
-3
7.0x10
-3
6.5x10
-3
6.0x10
-3
5.5x10
-3
5.0x10
-3
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
Resistivity (Ω cm)
4.5x10
-3
4.0x10
-3
3.5x10
-3
3.0x10
-3
2.5x10
-3
2.0x10
-3
1.5x10
-3
1.0x10
-4
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.25: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% MgO MT composite bulk.
99
-3
7.0x10
-3
6.0x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-3
Resistivity (Ω cm)
5.0x10
-3
4.0x10
-3
3.0x10
-3
2.0x10
-3
1.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.26: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% MgO composite bulk.
100
-2
4.0x10
-2
3.5x10
-2
Resistivity (Ω cm)
3.0x10
-2
2.5x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-2
2.0x10
-2
1.5x10
-2
1.0x10
-3
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.27: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 20wt% MgO MT composite bulk.
101
-1
1.6x10
-1
1.4x10
Resistance (Ohm cm)
-1
1.2x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-1
1.0x10
-2
8.0x10
-2
6.0x10
-2
4.0x10
-2
2.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.28: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 35wt% MgO MT composite bulk.
The broadening of the resistive transition for polycrystalline HTS in applied dc
fields has been explained previously, with excess resistivity (ρe) being suggested as
a better way of comparing the MR effect in Bi-2212 composites. Excess resistivity
as defined by equation 4-1 has been calculated for Bi-2212 and Bi-2212/MgO
composites with different concentrations of MgO additions (Figures 4.29 – 4.34). A
comparison between the five Bi-2212/MgO composites and pure Bi-2212 reveals
that the excess resistivity or the size of the MR effect is two orders of magnitude
102
greater in the case of 20wt% MgO compared to pure Bi-2212, similar to the Bi2212/USr2CaO6 composite.
As expected, the magnetoresistance of the Bi-2212/MgO composite increases with
an increased area of grain boundaries. Normally the supercurrent within a bulk
superconductor will pass through the path of least resistance and distance. As the
concentration of grain boundaries increases the ability of the supercurrent to avoid
the introduced barriers decreases and the current is forced through the network of
SIS junctions.
This behavior is responsible for the broadening of the excess resistivity peaks, since
the supercurrent will be forced through the MgO filled grain boundaries even at
lower temperatures as the concentration of MgO is increased. Once the MgO starts
to form agglomerations, however, the behavior of the resistivity is reversed. Grain
boundaries will become less common leading to the number of current paths that
avoid the MgO increasing, particularly at low temperatures. This can be seen in the
sharpening of the excess resistivity peaks and the reduction in the difference
between applied fields. This is why the Bi-2212 + 15wt% MgO composite looks like
it does. Texturing forces the MgO into grain boundaries and without texturing the
MgO addition forms agglomerations more easily within the composite.
103
Resistivity (Ω cm)
9.0x10
-4
8.0x10
-4
7.0x10
-4
6.0x10
-4
5.0x10
-4
4.0x10
-4
3.0x10
-4
2.0x10
-4
1.0x10
-4
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.29: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 MT bulk.
104
0.0035
Excess Resistivity (Ω cm)
0.0030
0.1T
0.5T
1T
2T
3T
5T
7T
9T
0.0025
0.0020
0.0015
0.0010
0.0005
0.0000
-0.0005
0
20
40
60
80
100
Temperature (K)
Figure 4.30: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 6.6wt% MgO MT composite bulk.
105
-3
4.0x10
-3
0.1T
0.5T
1T
2T
3T
5T
7T
9T
3.5x10
-3
Excess Resistivity (Ω cm)
3.0x10
-3
2.5x10
-3
2.0x10
-3
1.5x10
-3
1.0x10
-4
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.31: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 15wt% MgO MT composite bulk.
106
-3
4.5x10
-3
4.0x10
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-3
3.5x10
-3
Resistivity (Ω cm)
3.0x10
-3
2.5x10
-3
2.0x10
-3
1.5x10
-3
1.0x10
-4
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.32: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 15wt% MgO composite bulk.
107
Excess Resistivity (Ω cm)
2.50x10
-2
2.25x10
-2
2.00x10
-2
1.75x10
-2
1.50x10
-2
1.25x10
-2
1.00x10
-2
7.50x10
-3
5.00x10
-3
2.50x10
-3
0.1T
0.5T
1T
2T
3T
5T
7T
9T
0.00
0
20
40
60
80
100
Temperature (K)
Figure 4.33: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 20wt% MgO MT composite bulk.
108
-2
6.0x10
-2
5.5x10
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-2
5.0x10
-2
4.5x10
-2
Resistance (Ω cm)
4.0x10
-2
3.5x10
-2
3.0x10
-2
2.5x10
-2
2.0x10
-2
1.5x10
-2
1.0x10
-3
5.0x10
0.0
-3
-5.0x10
0
20
40
60
80
100
Temperature (K)
Figure 4.34: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 35wt% MgO MT composite bulk.
For a potential magnetic sensing device the relative variation of resistivity (ρr) is an
important characteristic, which has been defined earlier with equation 4-2.
Expressed as a percentage, the relative variation of resistivity at 77K for Bi-2212,
Bi-2212 + 6wt% MgO, 15wt% MgO (MT and untextured), 20wt% MgO and 35wt%
MgO is presented in Figure 4.35.
By 1T ρr for Bi-2212 + 15wt% MgO MT bulk and Bi-2212 + 20wt% MgO MT bulk
has dropped to just over half its original value. This steep slope at low fields is an
109
important feature for a magnetic sensing device because it is indicative of
sensitivity. The steeper the slope of ρr is, the greater the change in resistance with
temperature, allowing for the measurements of smaller increments of magnetic field.
2
8x10
Pure Bi-2212
6.6wt% MgO addition
15wt% MgO addition
15wt% MgO Addition - Not Textured
20wt% MgO addition
35wt% MgO addition
2
Relative Resistance [ρr] (%)
7x10
2
6x10
2
5x10
2
4x10
2
3x10
2
2x10
2
1x10
0
0
2
4
6
8
10
H (77K)
Figure 4.35: The field dependence of the relative variation in (ρr) at 77K, for Bi2212 + 0, 6.6wt% MgO, 15wt% MgO, 20wt% MgO and 35wt% MgO MT bulk,
respectively. The ρr for untextured Bi-2212 + 15wt% MgO is also included for
comparison.
Here we can see more clearly what is happening within the composite material as the
MgO concentration is increased through the maximum shown in Figure 4.19. As the
number of grain boundaries increases the supercurrent is forced through a growing
network of SIS junctions corresponding to its increasing sensitivity to applied
110
magnetic field. As the number of grain boundaries reaches the maximum number
within the composite the magnetoresistance also reaches a maximum because of the
high sensitivity and ease with which an applied magnetic field can decouple a weak
link which formerly passed supercurrent and change the resistance of the material as
the supercurrent is forced through another current path within the composite.
This means that it would be simple enough once calibrated to use this material’s
known resistance to establish an unknown magnetic field. To demonstrate the
concept of Bi-2212 + 20wt% MT MgO as a more viable composite material for a
superconducting magnetic field sensor a prototype was constructed. The dimensions
of the active part were 8.00mm x 2.00mm with a thickness of 0.75mm and four
contacts attached using a low resistance silver paste. With the applied magnetic
field perpendicular to the ab plane of the sensor the variation in resistivity was
measured by PPMS while the applied magnetic field was cycled at 77K. The lack of
hysteretic behavior in Figure 4.36 should be noted and considered when designing a
superconducting magnetic field sensor.
111
-3
8.5x10
-3
T=77K
I=.5mA
Sensor Resistivity (Ω cm)
8.0x10
-3
7.5x10
Field increase
Field decrease
-3
7.0x10
-3
6.5x10
-3
6.0x10
-3
5.5x10
-3
5.0x10
0
200
400
600
800
1000
Applied Magnetic Feild (mT)
Figure 4.36: The change of sensor resistivity with cycled applied magnetic field
for Bi-2212 + 20wt% MgO MT at 77K.
4.3 Bi-2212/CeO2 Composite
4.3.1 Materials Characterization – XRD
The X-ray analysis on the Bi-2212 MT disks with various CeO2 concentrations is
presented in Figure 4.37. By comparing the spectra of the MT pellets it can be seen
that each has a single phase Bi-2212 structure and is well aligned in the (00l)
direction with the added CeO2 phase identified as a separate phase against the host
112
matrix. This indicates that the superconductor coexists with the insulating material
and has a limited impact on the single phase Bi-2212 structure.
However, as the level of CeO2 addition increases there is a tendency for the Bi-2212
peaks to drift, indicating that some CeO2 within the composite is moving into the Bi2212 lattice. It has been previously reported that the Ce atom replaces the Ca in the
Bi-2212 lattice 99, and this behavior has a visible effect on the lattice parameters of
the Bi-2212 phase. The c parameter is seen to decrease monotonically with
increasing CeO2 additions while the a parameter increases slightly (Table 4.3), in
agreement with earlier reports99. The increasing a parameter corresponds to the
larger Ce atom moving into the lattice while the decreasing c parameter
demonstrates a decrease in intercalated oxygen, which has serious implications for
conductivity within the Bi-2212/CeO2 composite.
113
I (cps)
I (cps)
008
6000
5000
4000
3000
2000
1000
0
0010
0012
0016
006
002
2500 0
2000
1500
1000
500
0
10
0
10
20
30
Bi-2212
CeO2
*
40
0020
50
60
70
Bi-2212 + 6.6wt% CeO2
20
30
40
50
60
70
I (cps)
3000
2000
*
1000
0
0
10
20
Bi-2212 + 15wt% CeO2
*
30
40
50
60
70
Angle (2θ)
Figure 4.37: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of CeO2 from 0wt%-15wt%.
Table 4.3: Typical lattice parameter values deduced experimentally for the Bi2212 + CeO2 composites.
wt% CeO2
0
7
15
c (Ẳ)
30.77900
30.7225
30.6915
a (Ẳ)
5.41000
5.43
5.444
Looking at the FWHM in degrees (Fig. 4.38) of the 008 and 0010 peaks gives an
insight into the composite out of plane alignment which can then be compared to
114
that of other compositions. As expected the out of plane alignment decreases with
increasing non-superconducting addition into the Bi-2212 + CeO2 composite.
The FWHM of the (008) and (00 10) Bi-2212 peaks is a measure of the average size
of the Bi-2212 grains within the Bi-2212/ CeO2 composite. It can be seen that by
increasing the concentration of added CeO2 in the composite the Bi-2212 crystallites
are decreasing in size, indicative of the increasing number of grain boundaries. As
the number of grain boundaries increases a corresponding increase in
magnetoresistance is expected as the supercurrent is forced to tunnel through this
increased number of boundaries.
115
12
10
FWHM (theta -s)
8
008
6
0010
4
2
0
0
2
4
6
8
10
12
14
16
Concentration of CeO2 addition
Figure 4.38: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of CeO2 from 0wt%-15wt%.
4.3.2 Materials Characterization – DTA
With one of the highest free energies of formation, tetravalent cerium oxide (Ce(IV)
– O(II)) is the most stable oxide of cerium meaning that the reaction is favorable and
the compound stable. It should be noted though that CeO2 can exhibit large
deviations from the above stoichiometry with CeO2-x (with x up to 0.3). Past
approximately 600oC CeO2 acts like a novel solid electrolyte with good oxide ion
116
conductivity. This has ramifications for the formation of the Bi-2212 structure since
the diffusion of oxygen is fast while cation diffusion is slow.
It is generally recognized that during partial melting Bi-2212 melts incongruently,
forming a liquid and other non-superconducting solid phases. On cooling the Bi2212 phase forms jointly from the liquid and solid phases. The main solid phases
during this process are dependant on the atmosphere with Sr8Ca6Cu24Ox (14:24)
forming in an O2 atmosphere whereas an entire host of other non-superconducting
phases forms in an air atmosphere. With the increased availability of oxygen ions in
the material during the formation period courtesy of the added CeO2, an O2
atmosphere would be simulated such that the main non-superconducting phase
produced during partial melting would be 14:24. The initial endothermal peaks
(shown in the DTA results in Figure 4.39) below the melting temperature of the Bi2212 + CeO2 powder mix, which is indicated by the secondary endothermal peaks
(also in Fig. 4.39), seem to support this view.
While this is advantageous for the formation of a well-aligned single Bi-2212 phase
the matter is complicated by the diffusion of Ce (IV) and replacement of Ca (II) in
the lattice, which is accompanied by a decrease in intercalated oxygen ions. As
mentioned earlier the oxygen content of the cuprate is related to the physical
properties and structure of the superconducting material and is directly related to the
cell lattice parameter c. Specifically, the amount of intercalated oxygen may have a
direct impact on the carrier concentration within the system.
117
Bi-2212
Bi-2212 + 6.6wt% CeO2
Bi-2212 + 15wt% CeO2
20
15
Heat Flow (arb units)
10
5
0
-5
-10
700
720
740
760
780
800
820
840
860
880
900
920
940
Temperature (C)
Figure 4.39: Differential thermal analysis of the Bi-2212 powder mixtures with
MgO concentrations from 0wt%-15wt%. The endothermal peaks are offset
against an arbitrary unit on the vertical scale.
4.3.3 Magnetic Properties – PPMS
The mixing of such a significant amount of CeO2 into Bi-2212 completely destroys
the percolation superconducting backbone in the CuO2 planes by forming grains of
non-superconducting material within the composite, which completely block all
118
supercurrent. Even so, there remain some isolated superconducting grains as shown
by the magnetic measurement in Figure 4 and the very small value of M’.
Evidence of so few superconducting grains demonstrate quite clearly that not only
have the artificial boundaries of CeO2 merged into agglomerates but the replacement
of Ca within so many sites of the Bi-2212 lattice has completely destroyed
superconductivity within the sample.
0.0000000
M' (emu)
-0.0000005
-0.0000010
-0.0000015
-0.0000020
-0.0000025
0
10
20
30
40
50
60
70
80
90
100
110
Temperature (K)
Figure 4.40: Magnetic Tc of Bi-2212 + 15wt% MT composite bulk
119
4.3.4 Transport Properties – 4 probe
The dependencies of resistivity on temperature and applied fields of 0T – 9T for the
MT bulk pellets of Bi-2212 and Bi-2212 + CeO2 composites are displayed in Figures
4.41-4.43. It can be seen that for the pure Bi-2212 MT pellet the temperature and
field dependence of the resistivity follow the expected trend as the temperature falls
from 100K, with a decrease demonstrating metallic behavior followed by a drop at
the critical temperature of approx. 93K. The temperature of the zero value of
resistivity, (ρ≤10-6 Ω cm) decreases with increasing applied field, and the transition
becomes broadened in high fields.
The Bi-2212 + CeO2 MT composite bulk does not display this behavior at all, rather
it displays an insulator like behavior where the resistance increases with decreasing
temperature with no magnetoresistive effect demonstrated to 9T. This behavior
matches that of a typical insulating material with localized charge carriers. The lack
of influence by the applied magnetic field up to 9T demonstrates that the charge
carriers are localized. For there to be a lack of magnetoresistance as shown in
Figures 4.42-4.43 when fields of up to 9T are applied perpendicular to the current
then the samples are almost certainly very good insulators with very large energy
gaps.
120
This is what happens when the agglomerations of an additive form and become
numerous enough to block supercurrent throughout the material. At this level of
addition the CeO2 also destroys superconductivity within the composite material.
-3
1.0x10
-4
9.0x10
0T
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
-4
8.0x10
-4
Resistivity (Ω cm)
7.0x10
-4
6.0x10
-4
5.0x10
-4
4.0x10
-4
3.0x10
-4
2.0x10
-4
1.0x10
0.0
-4
-1.0x10
0
20
40
60
80
100
Temperature (K)
Figure 4.41: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk.
121
2
4.5x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
2
4.0x10
2
3.5x10
Resistivity (Ω cm)
2
3.0x10
2
2.5x10
2
2.0x10
2
1.5x10
2
1.0x10
1
5.0x10
0.0
1
-5.0x10
0
50
100
150
200
250
300
Temperature (K)
Figure 4.42: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% CeO2 MT composite bulk.
Applied field ┴ Current density.
122
2
3.50x10
2
3.25x10
2
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
3.00x10
2
2.75x10
2
Resistivity (Ω cm)
2.50x10
2
2.25x10
2
2.00x10
2
1.75x10
2
1.50x10
2
1.25x10
2
1.00x10
1
7.50x10
1
5.00x10
1
2.50x10
0.00
1
-2.50x10
-25
0
25
50
75
100 125 150 175 200 225 250 275 300 325
Temperature (K)
Figure 4.43: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% CeO2 MT composite bulk.
Applied field ┴ Current density.
To test this theory log10R vs. 1/T plots were plotted to experimentally determine the
energy gaps of these composites. If the Bi-2212/CeO2 composites were perfect
insulators then the log10R vs. 1/T plots seen in Figures 4.44 and 4.45 would have
shown a linear dependence with the slope used to estimate a value for the energy gap
(Eg, where Eg = Log10ρ/CT, where C is a constant). Instead we see a quadratic
dependence, which seems to be the result of linear addition of multiple slopes. What
123
we see then in the Bi-2212/CeO2 composites is a very dirty insulating material with
lots of impurities obscuring the band gap.
As expected the Bi-2212/CeO2 becomes more insulating with an increased
concentration of CeO2 addition. The initial gradients are large corresponding to large
Eg, thus the insulating behavior gets stronger with increased CeO2 concentration.
3.0
Gradient = 90
2.5
Log10R (Ω cm)
2.0
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
1.5
1.0
0.5
0.0
-0.5
0.00
0.05
0.10
0.15
0.20
-1
1/T (K )
Figure 4.44: The dependence of Log10 ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% CeO2 MT composite bulk.
124
2.5
Gradient = 105
2.0
Log10R (Ω cm)
1.5
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
1.0
0.5
0.0
-0.5
0.00
0.05
0.10
0.15
0.20
-1
1/T (K )
Figure 4.45: The dependence of Log10 ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% CeO2 MT composite bulk.
4.4 Bi-2212/Ag Composites
4.4.1 Material Characterization – XRD
The X-ray analysis on the Bi-2212 MT disks with various combinations of Ag
concentrations is presented Figure 4.46. By comparing the spectra of the MT pellets
it can be seen that each is well aligned in the (00l) direction with the added Ag phase
125
peaks being masked by Bi-2212 peaks much as with the other composites produced.
There exists no drifting of the Bi-2212 peaks, indicating that no Ag is being
incorporated into the lattice. This is demonstrated by the experimentally determined
lattice parameters in Table 4.5. The small change in c is a result of a change in
oxygen intercalation into the lattice rather than any direct affect from the Ag ions.
This extremely important result means that the Ag maintains its identity within the
Bi-2212/Ag composite despite the high temperature and pressures involved in
I (cps)
I (cps)
I (cps)
processing.
008 0010
6000
5000
4000
3000
2000
1000
0
0012
*
006
600 0
500
400
300
200
100
0
10
0
10
20
30
40
50
60
Bi-2212
Ag
70
Bi-2212 + 6.6wt% Ag
20
30
40
50
60
70
4000
3000
2000
1000
0
Bi-2212 + 15wt% Ag
0
10
20
30
40
50
60
70
Angle (2θ)
126
Figure 4.46: Comparison of x-ray spectra of Bi-2212 MT pellets with
concentrations of Ag from 0wt%-35wt%. Also included is the spectrum of an
untextured Bi-2212 + 15wt% Ag pellet.
Table 4.5: Typical lattice parameter values deduced experimentally for the Bi2212 + Ag composites.
wt% Ag
0
6.6
15
c (Ẳ)
30.77900
30.88
30.7789
a (Ẳ)
5.41000
5.41
5.41
Looking at the FWHM in degrees (Fig. 4.47) of the 008 and 0010 peaks gives an
insight into the composite out of plane alignment, which can then be compared to
the results of other concentrations of active components. This measurement confirms
the introduction of an increasing number of artificial grain boundaries through the
addition of Ag to the Bi-2212 powder.
The FWHM of the (008) and (00 10) Bi-2212 peaks are a measure of the average
size of the Bi-2212 grains within the Bi-2212/Ag composite. It can be seen that by
increasing the amount of added Ag powder to the composite the Bi-2212 crystallites
are decreasing in size, indicative of an increasing number of grain boundaries. As
the number of grain boundaries increases a corresponding increase in
127
magnetoresistance is expected as the supercurrent is forced to tunnel through this
larger number of boundaries.
14
12
FWHM (theta - s)
10
8
008
0010
6
4
2
0
0
2
4
6
8
10
12
14
16
Concentration of Ag addition (wt%)
Figure 4.47: Comparison of FWHM of the (008) and (00 10) peaks of Bi-2212
MT pellets with concentrations of Ag from 0wt%-15wt%.
4.4.2 Materials Characterization – DTA
Figure 4.48 shows the DTA results for the powder mixtures of ready made Bi-2212
and Ag additions in concentrations from 0wt%-15wt%. The endothermal peaks
128
represent the melting point of the Bi-2212/Ag composites, with shifting of the DTA
melting curve demonstrating the influence that changing Ag concentrations has on
the materials characteristics of Bi-2212. Earlier reports94 confirm the results shown
in Figure 4.48, and it is clear that the metal Ag is dissolving in the molten Bi-2212
phase during melting, thus lowering the melting point. The Ag effect on the melting
point seems to have saturated at 6.6wt% additions.
As the concentration of Ag is increased from 0 wt%- 15 wt% the change in the
melting profile becomes more prominent. The gradient of the heat flow preceding
the endothermal spike, which signifies the melting of the powder mixture, decreases
such that at 15wt% addition it approaches 0. The melting of a material originates in a
single spot spreading throughout the material until it is entirely melted. When Ag
particulates are dispersed throughout the powder, they provide initiation sites for the
melting of the material, as indicated by the decreased melting point. As the
concentration of Ag particulates increases the number of sites increases such that
melting starts to occur simultaneously throughout the sample.
As the Ag is dissolved within the liquid phase of Bi-2212 it lowers the temperature
of formation during cooling, as demonstrated by the lowered melting point (Fig.
4.48), thus giving a wider temperature margin to align the Bi-2212 grains and
control the microstructure in the molten stage and allowing for a decrease in nonsuperconducting phases.
129
Bi-2212
Bi-2212 + 6.6wt% Ag
Bi-2212 + 15wt% Ag
15
Heat Flow (arb units)
10
5
0
-5
-10
-15
700
720
740
760
780
800
820
840
860
880
900
920
940
Temperature (C)
Figure 4.48: Differential thermal analysis of the Bi-2212 powder mixtures with
Ag concentrations from 0wt%-15wt%. The endothermal peaks are offset
against arbitrary units on the Y-axis.
4.4.3 Transport Properties – 4 probe
The dependencies of resistivity on temperature and applied fields of 0T – 9T for the
MT bulk pellets of Bi-2212 and Bi-2212 + Ag composites are displayed in Figures
4.49-4.51. It can be seen that for the pure Bi-2212 MT pellet the temperature and
field dependence of the resistivity follows the expected trend as the temperature falls
from 100K, with a decrease demonstrating metallic behavior followed by a drop at
130
the critical temperature of approx. 93K. The temperature of the zero value of
resistivity, (ρ≤10-6 Ω cm) decreases with increasing applied field and the transition
becomes broadened in high fields.
The addition of Ag lowers the room temperature resistivity by an order of magnitude
but does not alter its temperature dependence. Although the onset of
superconductivity for the Bi-2212 + Ag composites is the same as for the pure Bi2212 pellets, the width of the normal-superconducting transition broadens slightly
with increasing Ag additions as expected by the increasing number of grain
boundaries. With increasing applied magnetic field this transition is seen to broaden
even further. The broadening is not nearly as large as with the MgO or the USr2CaO6
addition composites however.
An increase in Bi-2212 grain coupling occurs, combined with the fact that the metal
barrier material between the grains has significantly lowered the resistance of the
bulk material as can be seen in Figures 4.49-4.51. Cu and Ag rich nonsuperconducting dendrites forming near the embedded metal Ag particles offset
these improvements in coupling and grain alignment such that the out of plane
alignment and superconducting transition slowly degrade with increasing Ag
concentrations rather than experiencing the dramatic improvement or worsening that
may have been expected.
131
-3
1.0x10
-4
9.0x10
0T
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
-4
8.0x10
-4
Resistivity (Ω cm)
7.0x10
-4
6.0x10
-4
5.0x10
-4
4.0x10
-4
3.0x10
-4
2.0x10
-4
1.0x10
0.0
-4
-1.0x10
0
20
40
60
80
100
Temperature (K)
Figure 4.49: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 MT bulk.
132
-4
4.0x10
-4
3.5x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-4
Resistivity (Ω cm)
3.0x10
-4
2.5x10
-4
2.0x10
-4
1.5x10
-4
1.0x10
-5
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.50: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 6.6wt% Ag MT composite bulk.
133
-4
2.0x10
-4
1.8x10
-4
1.6x10
0T
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-4
Resistivity (Ω cm)
1.4x10
-4
1.2x10
-4
1.0x10
-5
8.0x10
-5
6.0x10
-5
4.0x10
-5
2.0x10
0.0
-5
-2.0x10
0
20
40
60
80
100
Temperature (K)
Figure 4.51: Dependence of resistivity ρ(T, H) on temperature and applied
magnetic fields of 0T - 9T for Bi-2212 + 15wt% Ag MT composite bulk.
The broadening of the resistive transition for polycrystalline HTS in applied dc
fields has been explained previously with excess resistivity (ρe) being suggested as a
better way of comparing the MR effect in Bi-2212 composites. Excess resistivity has
been defined earlier with equation 4-1 and has been calculated for Bi-2212 and Bi2212 + Ag additions with the results shown in Figures 4.52 – 4.54. A comparison
between the two Bi-2212 + Ag composites and pure Bi-2212 reveals that the excess
resistivity or the size of the MR effect is far less in Bi-2212 + Ag composites
134
compared to pure Bi-2212, as seen in Figures 4.52-4.54. This is due to the decrease
in overall resistance of the composite through the addition of a metal contaminant.
-4
9.0x10
-4
0.1T
0.5T
1T
2T
3T
4T
5T
6T
7T
8T
9T
8.0x10
-4
7.0x10
-4
Resistivity (Ω cm)
6.0x10
-4
5.0x10
-4
4.0x10
-4
3.0x10
-4
2.0x10
-4
1.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.52: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 MT bulk.
135
-4
2.5x10
0.1T
0.5T
1T
2T
3T
5T
7T
9T
-4
Excess Resistivity (Ω cm)
2.0x10
-4
1.5x10
-4
1.0x10
-5
5.0x10
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.53: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 6.6wt% Ag MT composite bulk.
136
Resistivity (Ω cm)
1.2x10
-4
1.0x10
-4
8.0x10
-5
6.0x10
-5
4.0x10
-5
2.0x10
-5
0.1T
0.5T
1T
2T
3T
5T
7T
9T
0.0
0
20
40
60
80
100
Temperature (K)
Figure 4.54: Dependence of excess resistivity (ρe) on temperature and applied
magnetic fields from 0T – 9T for Bi-2212 + 15wt% Ag MT composite bulk.
It was expected that the resulting network of SNS junctions created by the new
artificial metallic grain boundaries would act to decrease the resistance of the bulk
composite; the question significant to this project was whether networks of SNS
were just as effective at creating the magnetoresistance effect as SIS boundaries. To
this effect the relative variation of resistivity (ρr) was calculated, ρr was defined
earlier in equation 4-2. Expressed as a percentage, the relative variation of resistivity
at 77K for Bi-2212, Bi-2212 + 6wt% Ag and 15wt% Ag is presented in Figure 4.55.
137
Bi-2212
6.6wt% Ag
15wt% Ag
6
Relative Resistance [ρr] (%)
5
4
3
2
1
0
0
2
4
H[77K] (T)
6
8
10
Figure 4.55: The field dependence of the relative variation in (ρr) at 77K, for Bi2212 + 0, 6.6wt% Ag and 15wt% Ag MT composite bulk respectively.
The demonstrated sensitivity at low fields (<1T) is indicative of magnetoresistivity
generated by a network of Josephson junctions. As the applied field is increased it
becomes easier and easier to decouple a junction, creating an insurmountable barrier
through which to tunnel. Of interest within the Bi-2212/Ag composite, however, is
that even as the number of grain boundaries increases the relative resistance is seen
to decrease. This is a behavior that is opposite to that of SIS networks and can be
explained by the decreasing sample resistance with increasing grain boundaries and
Ag addition.
138
Even with a measurable relative resistance the Bi-2212/Ag composite would never
become a viable magnetoresistor with only a 6% increase in magnetoresistance.
4.5 Search for a suitable encapsulation material.
Just as important to the design of a commercially marketable magnetic field sensor
as the HTS composite core is the encapsulation material. It is this that allows the
potential sensor to be user friendly and impact resistant, and which extends the
lifetime of the sensor. Of vital importance to the selection of a suitable encapsulation
material is its resistance to thermal shocks, and to this end as mentioned previously
both the aluminum-filled epoxy and the polyurethane were repeatable cycled
between liquid nitrogen temperatures and room temperatures.
It was soon obvious that the relatively hard epoxy was not a suitable material. After
the first cycle substantial cracks developed throughout the bulk of the material with
the encapsulating material crumbling completely after only three cycles. Such a
short lifetime would be a substantial limiter to the commercial viability of a HTS
magnetic field sensor.
The polyurethane proved to be a much better encapsulation material with no visible
effect being observed after more than ten cycles. This was partially expected from
139
the natural ‘sponginess’ of the material ensuring that it has plenty of flexibility under
thermal expansion and contraction.
Chapter V: Conclusion
It has been clearly demonstrated through the addition of chemically compatible
insulators and metals to HTS Bi-2212 that a network of Josephson junctions is
created in the bulk material. The magnetoresistive behavior that results from
supercurrents passing through this network is promising for magnetic field sensing
applications. Differing behaviors between Bi-2212 composites and different degrees
of magnetoresistivity have made a study of this phenomenon necessary, with the Bi2212/MgO composite demonstrating it’s superiority in this area (Fig. 4.56). Here the
relative resistances, an important characteristic in magnetosensing, are compared.
140
Bi-2212
6.6wt% Ag
20wt% MgO
5.5wt% USr2CaO6
800
Relative Resistance [ρr] (%)
700
600
500
400
300
200
100
0
0
2
4
6
8
10
H [77K] (T)
Figure 4.56: The field dependence of the relative variation in (ρr) at 77K, for Bi2212, 6.6wt% Ag and 20wt% MgO and 5.5wt% USr2CaO6 MT composite bulk
respectively.
The maximum demonstrated relative resistance with each other type of addition are
compared with Bi-2212/MgO showing clear advantages for use as the active
component of a magnetic field sensor. At 77K the magnetoresistive effect is still
significant at <3T and over a temperature range of 10-90K. This can be compared to
the next closest candidate for use as a magnetoresistor, Bi-2212 + 5.5wt% USr2CaO6
with the magnetoresistance significantly <1T and effective over a temperature range
of 40-80K.
141
While it is clear from the demonstrated results that the maximum area of artificial
grain boundaries is introduced at a concentration of 20wt% MgO and hence ρe and ρr
are maximized, it should be noted that further optimization needs to be undertaken.
From the limited optimization undertaken within this project the optimal amount has
been narrowed to 17-20% MgO addition with no information on the effect of
optimized heat treatment included in this work.
While the search for an ideal encapsulation material for the Bi-2212 + 20wt% MgO
MT active component was started with a potential candidate determined, the search
was far too brief to be certain than an optimal material for use in a magnetic field
sensor was found.
Chapter VI: Sensor Fabrication
After the conclusion of the study of the magnetoresistance effect in Bi-2212 through
the characterization of four different Bi-2212 composites and the search for a
suitable encapsulating material two prototype HTS magnet field sensor were built
(Figure 4.57), using the most suitable Bi-2212 composites for the core, MgO and
USr2CaO6.
142
Figure 4.57: A schematic of the prototype superconducting magnetic field
sensor.
To test the feasibility of the Bi-2212/USr2CaO6 composite prototype it was placed
inside an electromagnet at 77K beside a semiconducting Hall probe (DTM-132). The
variation in sensor resistivity was measured in low magnetic fields using a current of
10mA and plotted as a function of applied field measured by the Hall probe (Figure
4.17). A small hysteretic behavior of approx 1% was observed at 400mT due to
measurement errors.
As is always the case in research projects however the desired testing and
characterizing of the Bi-2212/MgO composite prototype was not done due to time
constraints. Although the magnetoresistance of the HTS composite active core was
143
studied it is unknown what effect the encapsulating material will have on the
electrical behavior of the core.
With thermal cycling there will be an increase in pressure placed on the Ag contacts
and HTS composite core, accelerating the damaging effects of thermal cycling.
Further work needs to be done on the impact of encapsulating the active component.
Chapter VII: Conclusion
In this project the magnetoresistive effect in Bi-2212 was studied for the purposes of
using the HTS in cryogenic magnetic field sensing applications. It was determined
that introducing chemically compatible materials into the bulk Bi-2212 would
artificially create grain boundaries within the material by creating Josephson-like
barriers between the grains. This had the interesting effect of increasing the
magnetoresistive effect in the resulting Bi-2212 composite.
To this effect three chemically compatible non-magnetic insulators and one
chemically compatible non-magnetic metal were used at different concentrations to
create a series of Bi-2212 composites. Within the composites a network of
Josephson junction-like grain boundaries were created. Bi-2212/USr2CaO6 and Bi2212/MgO composites showed significant improvements in magnetoresistance. Bi2212/CeO2 composites showed a strong insulating behavior, and Bi-2212/Ag
composites showed small increases in magnetoresistance.
144
Bi-2212/MgO showed the most potential as the active component in any
commercially viable HTS magnetic field sensor, utilizing a Bi-2212 core with a ρr of
approximately 800%. Aside from any performance advantages Bi-2212/MgO
composites have over the other Bi-2212 composites, MgO is non-toxic, a clear
advantage over Bi-2212/USr2CaO6 composites.
Investigating a possible encapsulating material led to the building of a working
prototype. Polyurethane proved to be a useful material in which to encapsulate the
material making the sensor increasingly user friendly and impact resistant.
With a clear cost advantage over existing semiconductor magnetic field sensors and
ease of production a Bi-2212/MgO magnetic field sensor would be ideal to fill this
commercial
niche.
145
References
1
D.R. Baselt et al. 1998. "A Biosensor Based on Magnetoresistance Technology," Biosensors &
Bioelectronics, Vol 13:731-739.
2
Michael J. Caruso and Tamara Bratland, Honeywell, SSEC Carl H. Smith and Robert Schneider,
Nonvolatile Electronics, Inc (1998).
3
H. Njima, S. Tsuchimoto, S. Kataoka, Jpn. J. Appl. Phys. Part 1 27 (1988) 746.
4
S. Bandyopadhyay, M. Cahay, Cond-matt/0408137 v1.
5
A. C. Wright, K. Zhang, A. Erbil, Phys Rev. B. 44 (1991) 863.
6
I. Giaever [Nobel lecture], Rev. Mod. Phys., 46, 245 (1974)
7
B. D. Josephson, Phys. Lett. 1, 251 (1962).
8
M. I. Petrov, D. A. Balaev, K. A. Shaihutdinov, K. S. Alexsandrov, Semicond. Sci. Technol. 14
(2001) 798.
9
M. I. Petrov, D. A. Balaev, D. M. Gokhfeld, S. V. Ospischev, K. A. Shaihutdinov, , K. S.
Alexsandrov. Phisica C 314 (1999) 362.
10
M Ionescu, B Winton, T Silver, S X Dou and R Ramer. Appl. Phys. Lett. 84 No 26 (2004) 53355337.
11
D. A. Balaev, K. A. Shaihutdinov, S. I. Popov, D. M. Gokhfeld, M. I. Petrov. Physica C 408-410
(2004) 943-944.
12
Hott. R, Narlikar. A. V; High Temperature Superconductivity and Materials, Springer, Berlin 2004
13
J. G. Bednorz, R. A. Muller, Z. Phys. B 64 (1986) 189.
14
C. W. Chu, P. H. Hor, R. L. Meng, L. Goa, Z. J. Huang, Y. Q. Wang, Phys. Rev. Lett. 58 (1987)
405
15
J. Mannhart, P. Chaudhari, D. Dimos, C. C. Tsuei, T. R. McGuire Phys. Rev. Lett. 61 (1988) 2476.
16
R. B. Stephens, Cryogenics 29 (1989) 399.
17
J. Halbritter, Phys. Rev. B 48 (1993) 9735.
18
A. Umezawa, Y. Feng, H. S. Edelman, T. C. Willis, J. A. Parrell, D. C. Larbalestier, G. N. Riley
Jr., W. L. Carter, Physica C 2212 (1994) 378.
19
K. K. Likharev, Rev. Mod. Phys 51 (1979) 101.
20
B. D. Josephson, in Superconductivity, edited by R. D. Parkes (Marcel Dekker, New York 1969),
Vol. 1, p 423 ff.
21
Y. V. Fominov, N. M. Chtchelkatchev, A. A. Golubov. Cond.-matt. Non-monotonic critical
temperature in superconductor/ferromagnet bilayers.
22
J. Bardeen, L. N. Cooper and J. R. Schrieffer, Phys Rev 106, 162 and 108, 1175 (1957)
23
H. Hilgenkamp, J. Mannhart, VII-D, Rev. Mod. Phys. 74. (2002) 485.
24
W. Anderson, Basic notions of condensed matter physics (Benjamin New York 1984).
25
N. N. Bogoliubov, D. V. Sirkov, Quantenfelder (Physik-Verlag, Weinheim, 1984).
26
S. Reich, D. Veretnik, I. Felner, U. Yaron, J. Appl. Phys. 72 (1992) 4805-4811.
27
J. J. Calabrese, M. A. Dubson, J.C. Garland, J. Appl. Phys. 72 (1992) 2958-2963.
28
M. I. Petrov, D. A. Balaev, S. V. Ospishchev, B. P. Khrustalev, K. S. Aleksandrov, Physica C 282287 (1997) 2447-2448.
29
M. I. Petrov, D. A. Balaev, D. M. Gohfeld, S. V. Ospishchev, K. A. Shaihudtinov, K. S.
Aleksandrov, Physica C 314 (1999) 51-54.
30
J. Jung, M. A-K. M. Mohamed, I. Isaac, L. Freidrich, Phys. Rev. B. 49 (1994) 12188-12199.
31
G. Xiao, F. H. Streitz, M. Z. Cieplak, A. Bakhshai, A. Gavrin, C. L. Chien, Phys. Rev. B. 38 (1988)
776-779.
32
M. I. Petrov, D. A. Balaev, K. A. Shaihutdinov, K. S. Aleksandrov, Supercond. Sci. Technol. 14
(2001) 798-805.
33
J. K. Thomas, . Koshy, J. Kurian, Y. P. Yadava, A. D. Damodaran, J. Appl. Phys. 76 (1994) 23762379.
146
34
M. I. Petrov, D. A. Balaev, S. V. Ospishchev, B. P. Khrustalev, K. S. Aleksandrov, Physica C 282287 (1997) 2453-2454.
35
D. A. Balaev, K. A. Shaihutdinov, S. I Popkov, M. I. Petrov, Solid state communications 125
(2003) 281-285.
36
J. Koshy, K. V. Paulose, M. K. Jayaraj, A. D. Damodaran, Phys. Rev. B. 47 (1993) 15304-15307.
37
V. Y. Tarenkov, A. I. D’yachenko, A. V. Vasilenko, Phys. Solid state 36 (1994) 1197.
38
B. I. Smirnov, T. S. Orlova, Phys Solid State 36 (1994) 1883.
39
D. Berling, B. Loegel, A. Mehdaouli, S. Regneir, C. Cananoni, J. Marfaing, Supercond. Sci. Tech.
11 (1998) 1292.
40
H. Ohta, T. Matsui, Physica C 185-189 (1991) 2581-2582.
41
M. I. Petrov, D. A. Balaev, K. A. Shaihutdinov, K. S. Aleksandrov, Phys. Solid State, 41 (1999)
881.
42
D. Koelle, R. Kleiner, F. Ludwig, E. Dantsker, J. Clarke, Rev. Mod. Phys 71 (1999) 631.
43
V. Ambegaokar, A. Baratoff, Phys. Rev. Lett. 10, (1963) 486.
44
I. O. Kulik, I. K. Yanson, The Josephson Effect in Superconductive Tunneling Structures
(Izdatel’stvo Nauka, Mosco, 1970) (Israel Program for Scientific Translations, Jerusalem, 1972).
45
A. F. Andreev, Zh. Eksp. Teor. Fiz. 49 (195) 655 (Sov. Phys. JETP 22 (1966) 455).
46
A. Furusaki, H. Takayanagi, M. Tsukada, Phys. Rev. B. 45 (1992) 563; Solid state commun. 78
(1991) 299.
47
A. Furusaki, in Physics and applications of mesoscopic Josephson junctions, edited by H. Ohta and
C. Ishii (The Physical Society of Japan, Tokyo, 1999), p.79.
48
H. Takayanagi, in Physics and Applications of Mesoscopic Josephson Junctions (The Physical
Society of Japan, Tokyo, 1999), p233.
49
I. O. Kulik, Zh. Eksp. Teor. Fiz. 57 (1965) 655 [Sov. Phys. JETP 30 (1970) 944].
50
I. O. Kulik, Dr. Sci. Thesis (in Russian), B. Verkin Inst. Low Temp. Phys. And Engng., Kharkov,
1972; I. O. Kulik (unpublished).
51
C. Ishii, Prog. Theor. Phys. 44 (1970) 1525; 47 (1972) 1464.
52
J. Bardeen, J. L. Johnson, Phys. Rev. B. 5 (1972) 72.
53
A. V. Svidzinsky, T. N. Antsygina, E. N. Bratus, Zh. Eksp. Teor. Fiz. 61, (1971) 1612 [Sov. J. Low
Temp Phys. 10 (1972) 131].
54
U. Gunsenheimer, U. Schussler, R. Kummel, Phys. Rev. B. 49 (1994) 6111.
55
K. Schwidtal, R. D. Finnegan, Phys. Rev. B 2 (1970) 148.
56
R. C. Jaklevic, J. Lambe, J. E. Mercereau, A. H. Silver, Phys. Rev. 130 A 1628.
57
W. C. Danchi, J. B. Hansen, M. Octavio, F. Habbal, M. Tinkham, Phys. Rev. B. 30 (1984) 2503
58
B. D. Josephson. Rev. Mod. Phys. 36 (1964) 216.
59
Y-J. Kim, Cond-mat/0401374 v2.
60
S.-K. Yip, Cond-mat/0002395 v1.
61
A. Barone, G. Paterno, in Physics and applications of the Josephson effect (John Wiley and Sons,
inc, New York, 1982).
62
A. Davidson, M. Tinkham. Phys Rev B 13 (1976) 3261.
63
E. Bruneel, A. J. Ramirez-Cuesta, I. Van Driessche, S. Hoste. Phys Rev B. 61 (2000) 9176.
64
M. Prester. Phys Rev B. 54 (1996) 606.
65
M. Prester. Supercond. Sci. Tech. 11 (1998) 333.
66
N. Nadifi, A. Ouali, C. Grigorescu, H. Faqir, O. Monnereau, L. Tortet, G. Vacquier, C. Boulesteix,
Superconduct. Sci. Tech. 13 (2000) 1174-1179.
67
B. Zeimetz, B. A. Glowacki, J. E. Evetts. Cond. Matt. 0204082.
68
R. Flukiger, Invited lecture, M2S-HTSC-VII, Rio de Janeiro 2003.
69
B. A. Marinkovic, S. Xia, L. A. Saleh, M. Sens, E. T. Serra, R. R. de Avillez, F. Rizzo, Mater. Res.
5 (2) (2002) 179.
70
B. A. Marinkovic, S. Xia, E. T. Serra, H. Borges, E. Hering, F. Rizzo, Physica C 34 (2004) 408410.
71
R Caillard, V. Garnier, G. Desgardin. Physica C 340 (2002) 101-111.
72
J. Orenstein, A. J. Millis. Science 288 (2000) 468.
73
P. W. Anderson. Science 235 (1987) 1196.
147
74
J. M. Tarascon, Y. Le Page, P. Barboux, B. G. Bagley, L. H. Greene, W. R. Mc Kinnon, G. W.
Hull, M. Giroud, D. M. Hwang. Phys. Rev. B 37(1988) 9382.
75
S. A. Sunshine, T. Siegrist, L. F. Shneemayer, D. W. Murphy, R. J. Cava, B. Batlog, R. B. Van
Dover, R. M. Flemming, S. H. Glarum, S. Nakahara, R. Farrow, J. J. Krajewski, S. M. Zahurak, J. V.
Waszezak, J. H. Marshel, P. Marsh, L. W. Rapp Jr., W. F. Peck, Phys. Rev B 38 (1988) 893.
76
R. G. Buckly, J. L. Tallon, I. W. M. Brown, M. R. Presland, N. E. Flower, P. W. Gilberd, M.
Bowden, N. B. Milestone. Physica C 156 (1988) 629.
77
H. W. Zandbergen, W. A. Groen, F. C. Mijlhoff, G. Van Tendeloo, S. Amelinkx, Physica C156
(1988) 325.
78
Y. Le. Page, W. R. Mc Kinnon, J. M. Tarascon, P. Barboux. Phys Rev B. 40 (1989)
6810.
79
M. R. Presland, J. L. Tallon, R. G. Buckley, R. S. Liu, N. E. Flower, Physica C 176 (1991) 95.
80
F. Jean, G. Collin, M. Andrieux, N. Blanchard, J. F. Marucco. Physica C 339 (2000) 269-280.
81
D. E. Farrell, S. Bonham, J. Foster, Y. C. Chang, P.Z. Jiang, K. G. Vandervoort, D. L. Lam, V. G.
Kogan. Phys Rev. Lett. 63 (1989) 782
82
K. Okuda, S. Kawamata, S. Noguchi, N. Itoh, K. kadowaki. J. Phys Soc. Jpn. 60 (1991) 3226
83
J. C. Martinez, S. H. Brongersma, A. Koshelev, B. Ivlev, P. H. Kes, R. P. Griessen, D. G. de groot,
Z. Tarnavski, A. A. Menovsky. Phys. Rev. Lett. 69 (1992) 2276
84
D. Milliken, PhD Thesis (2004) 96.
85
B. Winton. Honors thesis, unpublished.
86
B. Soylu, N. Adamopoulos, D. M. Glowacka, J. E. Evetts. Appl. Phys. Lett. 60 (1992) 3183
87
S-L. Huang, D. Dew-Hughes, D. N. Zheng and R Jenkins. Supercond. Sci. Technol. 9 (1996) 368373
88
W. Wei, Y. Sun, J. Schwartz. IEEE. Trans. Appl. Supercond. 7 (1997) 1556
89
W. Wei, J. Schwartz, K. C. Goretta, U. Balachdran, A. Bhargava. Physica C 298 (1998) 368
90
K. Christova, A. Manov, J. Nyhus, U. Thisted, O. Herstad,. S. E. Foss, K. N. Haugen, K. Fossheim.
Journal of Alloys and Compounds 340 (2002) 1-5
91
L. F. Castro, R. Suryanarayanan, A. Das, E. Bacca, M. E. Gomez, W, Lopera, P. Prieto, A.
Kreisler, J.C. Martin. Solid State Communications 95 (12) (1995) 829-831.
92
S. Jin, R. C. Sherwood, T. H. Tiefel, G. W. Kammlott, R. A. Fastnacht, M. E. Davis, S. M.
Zahurak. Appl. Phys. Lett 52 (1988) 1628.
93
D. C. Shin, S-J. L. Kang, H-S. Chung, Physica C 340 (2000) 141-148.
94
M. Ionescu, S. X. Dou, M. Apperley, E. W. Collings, Supercond. Sci. Technol. 11 (1988) 1095.
T. Lang, D. Buhl, L. J. Gauckler, Supercond. Sci. Technol. 10 (1997) 311.
95
N. Nishiura, R. O. Suzuki, K. Ono, J. Am. Ceram. Soc. 81 (1998) 2181.
96
R Caillard, V Garnier, G Desgardin, Physica C 340 (2000) 101-111.
97
K. Konstantinov, S. Karbanov, Physica C 165 (1990) 170.
98
K. Konstantinov, S. Karbanov, Physica C 165 (1990) 170.
99
Li. B, Wu. B-M, M. Ausloos, Chin. Phys. Lett 21 (7) (2004) 1337-1339.
148