Parinov I.A. Microstructure and Properties of High-Temperature Superconductors

Подождите немного. Документ загружается.

108 3 Experimental Investigations of HTSC

Deformation was estimated as [776] ε =6dΔl/L

and deformation rate as

˙ε =6dV

,whereV

=Δl/t is the loading speed, d is the thickness tape,

L is the distance between supports. A test specimen is loaded consistently in

several sites. The AE signal activity was discontinuous in the form of single

bundles, characterizing non-uniformity of the fracture process or defect initi-

ation under loading, which was proper for brittle fracture.

First, a contribution in the AE activity from deformation of a silver sheath

was stated. For this, the tests of pure silver tapes with the same sizes and

conditions of loading, as for the tested Bi-2223/Ag tapes, have been carried

out. Moreover, it has been proved that loading device did not cause acoustic

noises. In order to conﬁrm a character of fracture and correlation of the AE

activity with the sample damage during loading, a comparison was carried

out between the obtained test data and the experimental results (obtained by

A. A. Polyanskii [856] by MOI method [859]) of critical current I

and crit-

ical current density J

in magnetic ﬁelds directed along c-axis of the tapes.

Magneto-optical images demonstrated pictures of the magnetic ﬂux arrange-

ments directly in regions loaded for two regimes: zero ﬁeld cooled (ZFC) and

ﬁeld cooled (FC). In ZFC, the sample was cooled below T

in the absence of a

ﬁeld, and then a ﬁeld was applied. In the resulting image (Figs. 3.7 and 3.8)

the sample was shielding the magnetic ﬁeld. For the FC regime, the sample

was cooled below T

in the presence of a magnetic ﬁeld, and then the ﬁeld

was turned oﬀ, so the sample was trapping magnetic ﬂux (Fig. 3.9).

The test results for monocore tapes #1 and #3 and also for multi-

ﬁlamentary tape #2 by using the AE and MOI methods are presented in

Tables 3.1 and 3.2 and Figs. 3.7–3.9. As it follows from Table 3.1 and Figs. 3.7

and 3.8, in total, there is a good correlation between results obtained by the

above two methods. Moreover, for sample #2, a tendency of the damage

increasing with increasing strain and deformation rate is obvious. The corre-

lation of the results in the case of sample #3 (see Tables 3.2 and Fig. 3.9) is

less obvious. Based on the AE tests of this sample, the following conclusions

are possible:

(1) At sites 2 and 3 with the smallest critical current, AE signals were also

observed. The site 3 (with the smallest critical current) was loaded in

the tests twice. It may be assumed that during ﬁrst loading (when electric

hindrances were observed), a sample cracking was initiated that was forced

at repeated loading.

(2) An absence of AE signals at the sites 1 and 4 (at the existence of cracks,

revealed by the MOI method) is related with the possible absence of a

good acoustic contact of the sample with the transducer or/and due to

the cracks that could be formed at the following accompaniment of the

sample (i.e., at marking of the sample, its straightening or/and in the

following measurements).

Additionally, the AE tests of the Bi-2223/Ag monocore tapes were carried

out with the following measurement of I

(at 77 K). At the testing of three

3.1 Experimental Methods of HTSC Investigations 109

1 mm

(a)

(b)

Fig. 3.7. Magneto-optical images of the monocore tape #1, ZFC regime (T = 13 K,

H = 600 Oe): (a) site 1 (two-side bending), six cracks are seen; (b)site2,3cracks

are seen; (c)site3,nocracks;and(d) site 4, weak defects [820]

samples with diﬀerent strain rates (one test for every sample) there is correla-

tion between changes of the strain rate and critical current. We obtained the

critical currents I

= 31, 24 and 18 A for deformable samples (for initial un-

strained sample I

= 39 A). They corresponded to strain rates: ˙ε/d =0.00030,

0.00049 and 0.00054(s mm)

−1

, normalized in the thickness, at visible absence

of the dependence on strain. Corresponding correlation was also observed for

acoustic emission.

110 3 Experimental Investigations of HTSC

1 mm

(a)

(c)

(d)

(b)

Fig. 3.8. Magneto-optical images of the multi-ﬁlamentary tape #2, ZFC regime

(T = 13 K, H = 1200 Oe): (a) initial unstrained tape; (b) site 1, 10–12 great cracks

are seen; (c)site2,nocracks;(d) site 3, some cracks are possible [820]

Thus, an increase of the strain rate caused a forcing of the AE activity,

accompanying greater microdamage and corresponding diminishes of critical

current. Hence, it is necessary to take into account the sample strain rate at

evaluation of the superconducting tape damage, additionally to the sample

deformation.

The sample strain is only taken into account at mechanical treatment of the

deformational behavior of structure-sensitive properties, usually.

3.1 Experimental Methods of HTSC Investigations 111

(a)

(b)

(c)

(d)

(e)

1 mm

Fig. 3.9. Magneto-optical images of the monocore tape #3, FC regime at

H=600Oe(T=12K,H=0Oe):(a) initial unstrained tape; (b) site 1, 4–5 great

cracks are seen; (c) site 2, 3–4 great cracks are seen; (d) site 3, 5 great cracks are

seen; (e) site 4, 4–5 great cracks are seen [820]

Based on the above tests, it could be concluded that AE method is suitable

to estimate the microdamage formation and propagation during Bi-2223/Ag

tape bending. An improvement of the test results may be related to using the

designed interference method [817] and the created optic-holographic device

for estimation of small displacements [705, 818].

112 3 Experimental Investigations of HTSC

Table 3.1. Test results for the samples #1 and #2 obtained by the AE and MOI methods

No. of

site

Δl

(mm)

(mm/s)

ε/d

(mm

−1

)

˙ε/d

(s mm)

−1

N (imp/s) ΣN

(imp)

ΣA

(×10

−7

(g) MOI data

#1,

site 1

3 0.1 0.0147 0.00049 – – – 5 –

#1,

site 1

1.5 0.06 0.0073 0.00029 80 (in 6 bundles) 800 0.2 5 6 cracks

#1,

site 2

2.4 0.15 0.0118 0.00074 130 (in 3 bundles) 400 – 5 3 cracks

#1,

site 3

2.9 0.2 0.0142 0.00098 – – – 5 No cracks

#1,

site 4

2.9 0.26 0.0142 0.00127 1500 (in 1 bundle) 3000 0.6 5 Weak defects

#2,

site 1

3 0.2 0.0147 0.00098 ≈ 1300 (in 1 bundle) 4000 3 80 10–12 great cracks

#2,

site 2

1.4 0.06 0.0069 0.00029 – – – 75 No cracks

#2,

site 3

2.9 0.08 0.0142 0.00039 150 (in 4 bundles) 600 0.2 80 some cracks are

possible

At second loading of considered site from the opposite side

3.1 Experimental Methods of HTSC Investigations 113

Table 3.2. Test results of AE and critical current for sample #3

Property Site 1 Site 2 Site 3 Site3

Site 4

Δl (mm) 6 6.5 4.3 6.5 6.5

(mm/s) 0.33 0.23 0.3 0.3 0.26

ε/d (mm

−1

) 0.0294 0.0319 0.0211 0.0319 0.0319

˙ε/d (s mm)

−1

0.00162 0.00113 0.00148 0.00148 0.00127

N (imp/s) – 300 (in 9 bundles) AE with electric

hindrances

400 (in 1 bundle) –

ΣN (imp) – 1000 400 –

ΣA(×10

−7

m) – 1 0.2 –

(g)1418182022

(A) 23 18 – 14 21

(kA/cm

) 15.6 12.2 – 9.5 14,2

AE data No AE There are active

defects

AE is not distin-

guished

Weak AE No AE

At second loading of considered site from the same side. For original no-deformed edge of the sample placed at the outer side of

supports at the ﬁrst loading, the measured critical current had the following values: I

=45AandJ

=30.6kA/cm

(at T =77K).

Maximum thickness of Bi-2223 core is equal to 60 μm, and S =1.47 × 10

−3

is the cross-section area

114 3 Experimental Investigations of HTSC

3.2 Intergranular Boundaries in HTSC

The properties of high-angle grain boundaries are believed to control the

macroscopic J

(H) characteristics of all polycrystalline HTSC. This control

occurs because most high-angle grain boundaries act like barriers to the cur-

rent and have electromagnetic properties, such as Josephson junction-like

properties [205]. On the other hand, in the melt-processed polycrystalline

YBCO, actual currents can penetrate through intergranular boundaries mis-

oriented up to 30

◦

and more. So, HTSC properties are connected closely with

distribution of the intergranular boundary misorientations [205, 649].

From both the high-ﬁeld ﬂux-pinning viewpoint and low-ﬁeld, Josephson

junction-based electronics viewpoint, there is strong motivation to develop a

detailed picture of the grain boundary structure and microstructure and to

describe their eﬀects on the electromagnetic properties of the grain bound-

aries. The superconducting coherence length, ξ, deﬁnes the grain boundary

thickness that may be penetrated by the supercurrent. In this case, barriers

with thickness up to a few coherence lengths can still show superconducting

coupling, albeit of reduced strength. At the same time, the grain bound-

aries in HTSC are defects with thickness in the 0.5–1.0 nm range, indeed,

approaching ξ [40, 582]. Super-short coherence length and great values of n

in the power dependence of E–J (where E is the electric ﬁeld intensity and

J is the current density) proper for HTSC [370, 372], sign, that the defects

with size of some nanometers can prevent supercurrent and create obstacles

with eﬀective size, that is considerably higher than nominal size of defect.

There are other planar defects, causing magnetic ﬂux pinning and supercur-

rent percolation in HTSC, namely (i) twinning in YBCO [288, 593, 622],

(ii) stacking faults [272, 677], (iii) colonies of low-angle c-axis intercrystalline

boundaries [272, 677, 1096], (iv) twist intergranular boundaries of the “brick

wall” type [104] and low-angle a-, b-axis boundaries of the “railway switch”

type [1186] in BSCCO, (v) overgrowth of Bi-2223 phase into Bi-2212 phase

[559], (vi) overgrowth of superconducting phase into silver sheath [887], (vii)

amorphous and normal (non-superconducting) phases [289], (viii) voids, mi-

crocracks and other crack-like defects, which are proper for all oxide supercon-

ductors [403, 743], (ix) macrodefects into Josephson junctions [1014], and also

various dislocation networks [634], discussed in detail below. Diﬀerent types of

planar defects causing the structure-sensitive properties of HTSC, are shown

in Fig. 3.10.

Polycrystalline YBCO samples can be divided into two broad classes, ac-

cording to microstructure features, namely (i) specimens heated above peri-

tectic temperature with the aim of obtaining oblong well-orientated grains

[14] and (ii) samples sintered with grain structure that is near to equal-axes

one [975]. The experimental studies relate to investigation of both individual

isolated grain boundaries and polycrystalline samples with averaged eﬀects of

great number of the intercrystalline boundaries. Both test types often sup-

plement each other. So, the study of high-angle boundaries in YBCO thin

3.2 Intergranular Boundaries in HTSC 115

(a)

(b)

(c)

µm

YBCO

200

130

221

50 nm

(221) Facet

Stacking Faults

Partial Dislocation

Fig. 3.10. Diﬀerent types of planar defects in HTSC: (a) {110} twinning boundaries

in the 10

◦

-bicrystal of YBCO (A and B are the macroscopic facets) [1082]; (b)

diﬀraction contrast image of the spatial dislocation network conﬁguration in grain

boundary (g = [200] is the diﬀraction vector), the dark and bright areas are associated

with long-range strain contrast in facet junctions [1079]; (c) schematic diagram,

showing the arrangement of dislocations in the (22

1) facet [1079]

116 3 Experimental Investigations of HTSC

20 nm

c-axis

(f)

(e)

(d)

g = 100

100 nm

Fig. 3.10. (d) screw dislocation in twist intergranular boundary [1131]; (e) lattice

fringe image, showing examples of a twist grain boundary (grains A and B), a

low-angle c-axis tilt colony boundary (colonies 1 and 2) and a low-angle ab-axis tilt

colony boundary (arrow) [242]; (f) twist intergranular boundaries of the “brick wall”

type [1186]

3.2 Intergranular Boundaries in HTSC 117

(g)

c-axis

Grain

The “railway switch” model with a

low-angle c-axis GB at the switch.

c-axis

The “railway switch” model with

a low-angle c-axis GB at the

switch. A special switch structure,

containing no (001) plane.

The “railway switch” model with

a low-angle c-axis GB at the

switch. A general switch structure,

containing part of the (001) plane.

(h)

(Sr, Ca)CuO

Bi- 2212 Bi-2223

20µm

Fig. 3.10. (g) diﬀerent models of the grain boundaries of the “railway switch” type

[1186]; (h) overgrowth of Bi-2223 phase into Bi-2212 phase [289]