PHYSICAL REVIE%' B VOLUME 51, NUMBER 18 1 MAY 1995-II

Field-induced granularity in a well-textured (Bi,pb)~sr~Ca~Cu3O» tape

G. C. Han, H. M. Han, Z. H. Wang, X. N. Liu, W. F. Yuan, and F. T. WangHigh Magnetic Field Laboratory, Institute of Plasma Physics, Academia Sinica, P.O. 1126, Hefei 230031, People's Republic of China

(Received 26 May 1994)

The magnetoresistivity of a well-textured (Bi,Pb),Sr&Ca2Cu30~ tape is studied experimentally for atemperature range very close to T, (T/T, )0.93) in a very low magnetic-field range (8 (0.1 T). In thisH-T region, the dissipation associated with magnetoresistance curves seems to arise from the granularityinduced by the field. As the temperature decreases, related Josephson junctions may well be coupled oreven become proximity-effect superconductors, thus the dissipation is dominated by the conventionalAux-creep process described by Anderson and Kim. This observation was further confirmed by the tem-perature dependence of the critical current density J,(T) at various magnetic fields. The J,(T) depen-dence seems to support the hypothesis that the Josephson coupling in the present sample is dominatedby superconductor-normal-superconductor proximity-e6'ect junctions.

INTRODUCTION

The transport properties of high-temperature super-conductors (HTS) in the presence of a magnetic field havebeen the subject of intense research, both experimentallyand theoretically over the past few years. ' In practicemost applications such as in superconducting magnets re-quire high current-carrying capacity often in a large mag-netic field, and the details of the resistive transition arethe important information for the design and the protec-tion of the magnets. Associated with the transition, twounique anomalous properties have been found. One is theexistence of an "irreversibility line, " reported by Muller,Takashige, and Bednorz, which had led to much debateover the variety of vortex states. The other is thesignificant broadening of the resistive transition in a mag-netic field. A lot of models have been proposed to ex-plain this effect in terms of Aux creep, Aux flow, Auctua-tions, and Josephson coupling, but none of these ap-proaches has been completely successful. The observa-tion of the Lorentz-force-independent dissipation in high-ly anisotropic oxides Bi2Sr2CaCuO (Bi-2212),TlzBa2CaCuzO, (T1-2212), and Y2Ba~Cus0, 6 (Y-124)(Refs. 9—12) casts some doubts on the dissipative mecha-nisms based on the framework of Aux motion. TheJosephson coupling model gives a natural explanation ofthe lack of any Lorentz-force dependence and is con-sistent with the broadened transition. But this modelseems not to be so successful in explaining the magne-toresistivity data, especially for temperatures close toT, . ' At temperatures near T„Raffy et al. ' have re-ported that the magnetoresistivity can be interpreted interms of two-dimensional superconducting fluctuationseven at temperatures well below T, . Their results seemto be well consistent with the scaling form p-ln(H), atsufficiently high fields. But an obvious deviation can alsobe found in the low-field region, typically below 1 T. Inan attempt to reveal the mechanism of the commonly ob-served "shoulder" behavior near T, in the broadenedresistive transition, more detailed studies, particularly atlow fields, are clearly needed. Up to now, most authors

have selected to characterize their sample with the mag-netic field applied parallel to the c axis. This type ofwork is of great importance to explore the vortex dynam-ics in high-T, superconductors. However, for most appli-cations, such as in superconducting magnets, the field isfrequently parallel to the film plane. In this paper, wepresent magnetoresistivity data at temperatures veryclose to T, (T/T, )0.93) in a very low field with the field

applied parallel to the film plane. The results are ana-lyzed in combination with the temperature dependence ofthe critical current density J,(T), and represent the dissi-pation arising from the field-induced granularity in thistemperature and field region.

EXPERIMENT

The sample for these experiments is a high-J, silver-clad Bi, sPbo4Sr2Ca2Cu30 (Bi-2223) film tape with di-1.8 0.4 2 2 3 ymensions of 14X2.5 X0.04 mm . The tape was preparedthrough a physical deposition method using the "powderin tube" technique. The details of the fabrication werereported elsewhere. ' From the x-ray diffraction patternfor the sample, no second phase was found and muchstronger (001) peaks were observed, e.g., the ratio of thestrengths of the (111) and (0010) peaks is 10%, demon-strating the high texture of the sample with a preferentialgrain orientation with the c axis perpendicular to the filmplane. ' The transport J, characteristics of the samplewere studied in previous work' and displayed practicalhigh-field properties around 50 K. The J, value was2.6X 10 A/cm at 77 K and 0 T and no weak-link efFectwas detected through commonly used transport measure-ments at 77 K. ' Magnetoresistance curves R(H) weremeasured by the dc four-probe technique and recordedthrough an X-Y recorder with a sensitivity of 5 pV/cm.The contacts were soldered to the silver sheath and wereso separated from one another as to ensure a uniformcurrent How through the region between the voltage con-tacts according to the calculation in Ref. 18. The fieldwas applied parallel to the film plane and perpendicularto the current direction unless otherwise stated. To per-form the R (H) measurement, an excitation current of 0.5

0163-1829/95/51(18)/12754{5}/$06.00 51 12 754 1995 The American Physical Society

12 75551

ment above T, and is as followsA was used and appi d 0r o o 1 d

tively.u i ized for

e thermom-an Rh-F, measurement s, respec- R 2% 2T 73 7

where R„and T are in p

b ofh

y. This

resistivitive intercept. H ur s

vi y measurementsowever, our s

(from 77 to 30o t e silveri out su er

o OK) of

its les 'to Eq. (1).a so showed

Suppo sin E 'e

e similar

g q. (1) is also e

ow 0 0 cmo t e contact

the supercondbt th 1

n e present 1 en uctor for te silver she

g23 fil 't lf R

1 111e relationship

RESULTS AND DDISCUSSION

Figures 1(a) and 1

curves at s(h) displa g

to e that the esistivity in

i respect to tn ent of the fie

o t."e current d'

orce is essentiairection. S'

d t d'an ing the diissip ation

, one has to cornetoresistivit oy of the Bi-

s eath. Assuata for the in

s h. umi gth tthlm, R, is

e normal reshl g h

sistance of

can be obt'

ained from the Re si vef

e R„(T) measure-

R =, =R R„/(R —R in m

where R is tIn the proc

is t e measured rresistance.

R, ( )frocess of fittin

ional relationshi ip cannot be sat' fi, we found that th e

e satisfied by the

FIELD-IND UCED GRANULULARITY IN A W

th

WELL-TEXTURW ED. . .

i00

60

0 0. 06

H (Y}

FIG. 1. (asilver-clad B'

agnetores'esistivity of a h h-

close tPape at tern ry

5 1064 K; curve 6: 106.68 1076 K.

() ' 'y a i h-J

tape at ternig -, silver-clad

for Hmperatures ver

Ill. Curve 1: 103.4 K. .9; c r 4: 104.9K;.9 K; curve 5:

8 107 1

.1 K; curve 9: 107.8 K.

100

60

0, 06

H (Y}

12 756 HAN, HAN, WANG, LIU, YUAN, AND WANG 51

scaling form p —ln(H) based on thermal critical fiuctua-tions, nor by the Josephson-coupling model. There-fore an alternative dissipation mechanism should be con-sidered. Notice that the Bi-2223 films are polycrystalline,albeit being well textured and with a high degree of c-axisorientation; Josephson coupling could be formed betweenthe grains. Furthermore, as pointed out by Daeumling,Seuntjens, and Larbalestier, ' the field suppresses the su-perconductivity in regions that are already weakened byeither point or extended microscopic defects, whichseems to be in good agreement with their magnetizationdata. Although the film showed a sharp resistive transi-tion at zero field, weak links or Josephson-coupled re-gions or grains could be induced as the field was in-creased. The grain boundaries may play a dominant rolein this field-induced granularity for the present sample,but the other formation mechanisms, such as from themicroscopic defects and the intrinsic weak coupling be-tween Cu-O layers in this highly anisotropic structure,could also be important. To see this eA'ect, experimentalstudies based on a high-quality single crystal are clearlynecessary.

The critical currents I, (H, T) of Josephson junctionsare inhomogeneous, depending on their origins. For aconstant current and at given temperatures, assumingthat the field H drives a number X of the junctions to theresistive-superconducting state, the number of the junc-tions with a resistance per unit field is N/H. Then thenumber dX of junctions produced in the field rangeH —(H+dH) should be proportional to N/H and dH,i.e., dN = n (N/H)dH, where n is a field-independentconstant. To simplify the complexity arising from therandom arrangement of the Josephson junctions, we con-sidered the simplest case that the tape is composed of aseries of junctions. If there are junctions connected inparallel, they can be regarded as an equivalent junction.In that case, the resistance of the superconductor R,would be proportional to X, and we have:

R, -N —[H —H, o( T)]", (3)

where H, o(T) is the field above which the magnetoresis-tivity appears and can be obtained from the experimentalresults. In Eq. (3) we use [H —H, o( T)] rather than H be-cause below H, o( T) all junctions remain well coupled andno resistance can be detected. To test Eq. (3), the magne-toresistivity data are presented again in logarithmic plotsof R, versus [H —H, o(T)] in Fig. 2. Power-law varia-tions of the magnetoresistivity can be clearly found.However, it should be pointed out that at temperatureswell below T, the power-law dependence can no longerbe found. The dissipation is instead dominated by athermally activated Aux-Row process. This observationimplies that the Josephson junctions in the present sam-ple are so well coupled that the eA'ect of the granularitywould become secondary at low temperatures, whichperhaps is the reason why this type of silver-clad tape orwire has a much higher J, than YBa2Cu30~ bulk materi-als.

The temperature dependence of the magnetoresistivitycan be analyzed by a commonly used method. As point-

6, 6

1n [8-Hco (Tj g) ]

6 ~ 6

FIG. 2. Magnetoresistance curves of the Bi-2223 film itselffor HlI in logarithmic plots of R, versus (H —H, o), where H, o

is the field above which the magnetoresistivity appears.

where the activation energy is assumed to have the formUo(H)(1 —T/T, )~. In the consideration of the field-

induced granularity, combining Eqs. (3) and (4), we ob-tain

R, =Ro(H —H, o)"exp[ —Uo(H)(1 —T/T, )~/kT] . (5)

If Uo(H) has the commonly used form of 1/H, thenEq. (5) can be written as

(I —T/T, )q/H = AkT ln[R (H H)"/R, ]—, (6)

where 2 is a constant. Because the right-hand side inEq. (6) is a relatively slowly varying function of T and Hfor T close to T„at the same resistance levels, Eq. (5)predicts an approximate scaling of the temperature widthof the transition as

(7)

where T, (R„O) is the critical temperature defined fromthe resistance level R, in the transition at zero field.

Figure 3 shows AT versus W for several resistance lev-els in ln-ln plots. The scaling form of Eq. (7) can be easilyfound from Fig. 3. One can find that the power q isaround 2, slightly decreasing at higher dissipative levels.

Further evidence of the field-induced granularity inthis silver-clad sample was found from measurements ofthe critical current density I, in a magnetic field. Thecritical current was determined with the voltage criterionof 1 pV/cm. Figure 4 indicates the temperature depen-dence of J, at three fields. At a fixed field, only when the

ed out by Kim et al. , based on the Josephson-couplingmodel, the resistivity can be scaled as the Arrhenius law

p=po exp( —U/kT) =po exp[ —Uo(H)(1 —T/T, ) i/kT],

(4)

51 FIELD-INDUCED GRANULARITY IN A WELL-TEXTURED. . . 12 757

7

0, 6 1. 0 1. 46

1n[T (R., 0) - Y(R., H) (K) j

FIG. 3. Temperature width versus magnetic field for severalfixed resistivity levels in logarithmic plots. From the top to thebottom, the lines are related to the resistance levels 50, 40, 30,20, and 10 pQ.

-t. 0

ln (l-Y jT. )

FIG. 5. Logarithmic plots of J,(T) in a temperature rangeclose to T, . curve a: 0.5 T; solid circles: 1.0T; curve b: 2.0T.

aH

I ~Qo

0 ~ 0(h) Q) ~ oo

o oo (a)

I- I I

O Pg+

I i %~a- ~50

7 (K}

$00

FIG. 4. Temperature dependence of critical current densityJ, ( T) at several magnetic fields; curve a: 0.5 T; solid circles: 1.0T; curve b: 2.0T.

temperature is well below T, can a finite J, be measured,as shown in Fig. 4. The boundary defined by J, =0 is stilla controversial issue, yielding many competing models.In our opinion, above the temperature below which J,appears, denoted as T, , the grains in the sample are in avortex-lattice melting state and the intergrain or in-tragrain Josephson junctions are decoupled. In the vicin-ity of T, , there may exist a transition of the vortex fromthe liquid to an Abrikosov vortex lattice. . From theAux-creep theory of Anderson and Kim, one would ex-pect a linear temperature dependence of J, . This linearJ,(T) dependence is indeed found at temperatures wellbelow T, . However, a nonlinear J, ( T) can also be clear-ly observed at temperatures close to T, in Fig. 4. Wethink this nonlinear J, (T) characteristic is due to theeffect of the field-induced granularity in the present

silver-clad sample. Above T, , all field-induced Joseph-son junctions are decoupled and lie in the resistive-superconducting state, thus resulting in a zero transportJ, . As the temperature decreases, the junctions may becoupled more and more strongly, so J, is increased rapid-ly. For different types of junctions, the temperaturedependence of their critical currents are also varied. Forsuperconductor-normal-superconductor (SNS) proximity-effect junctions, according to the theory of de Gennes,I, ( T)—(1—T/T, ) . But for an insulating boundary layer[superconductor-insulator-superconductor (SIS) junc-tion], the calculations by Ambegaokar and Baratoffindicate I, ( T) —(1—T/T, ). For a high-J, silver-clampedBi-2223 sample, we found J, —(1—T/T, )

~ for T nearT, and at zero field. '

To examine the J,(T) dependence, the J, data are re-plotted on logarithmic plots of J, versus (T —T, ) in atemperature region near T, in Fig. 5. The determina-tion of T, may be a little ambiguous. Here the T,values were taken as the temperatures above which J,tends to zero (below 50 A/cm ) and the best fit of thedata was to a power law of the type J, —(T, —T)

From Fig. 5, one finds that the power I is approxi-mately equal to 2. This results seems to suggest that thefield-induced Josephson coupling is dominated by SNSproximity-effect junctions. At low temperatures, thejunctions are strongly coupled and may finally becomeproximity-effect superconductors. Therefore the effectof such field-induced granularity would become secon-dary, and J, would be dominated by the Aux-creep mod-el, resulting in a linear J, ( T) dependence.

In summary, magnetoresistance curves were measuredin a temperature range very close to T, . The results seemto provide evidence that Josephson coupling can be in-duced by a magnetic field. The temperature dependenceof the transport J, at various magnetic fields indicated

12 758 HAN, HAN, WANG, LIU, YUAN, AND WANG 51

that the Josephson coupling was dominated by SNSproximity-effect junctions. At lower temperatures, thistype of junction may be strongly coupled or even becomeproximity-effect superconductors. Further experimentalstudies based on a high-quality single crystal in a similartemperature and Geld regime would be very interesting.

ACKNOWLEDGMENT

This work was supported by the National Center ofResearch and Development on Superconductivity of Chi-na under Grant No. J-A-4226. Thanks are also given toZhang Yong for assistance in the measurements.

'FAX: 0086-551-5591310.A. Freimuth, in Selected Topics in Superconductivity, edited by

M. S. Multani and L. C. Gupta, Frontiers in Solid State Sci-ences Series (World Scientific, Singapore, 1993).

2K. A. Muller, M. Takashige, and J. G. Bednorz, Phys. Rev.Lett. 5S, 1143 (1987).

B. J. Bishop, P. L. Gammel, D. A. Huse, and C. A. Murray,Science 255, 165 (1992}.

4For example, T. T. M. Palstra, B. Batlogg, L. F. Schneemeyer,and J. V. Waszczak, Phys. Rev. Lett. 61, 1662 (1988).

5T. T. M. Palstra, B. Batlogg, R. B. van Dover, L. F.Schneemeyer, and J. V. Waszczak, Appl. Phys. Lett. 54, 763(1989).

M. Tinkham, Phys. Rev. Lett. 61, 1658 (1988).7K. Kitazawa, S. Kambe, and M. Naito, in Strong Correlations

and Superconductivity, Proceedings of the IBM Japan Inter-national Symposium, edited by H. Fukuyama, M. Maekawa,and A. P. Malozemoff (Springer-Verlag, Heidelberg, 1989), p.361.

D. H. Kim, K. F. Gray, R. T. Kampwirth, and D. M. Mckay,Phys. Rev. B 42, 6249 (1990}.

Y. Iye, S. Nakamura, and T. Tamegai, Physica C 159, 433(1989).

OH. Iwasaki, N. Kobayashi, M. Kikuchi, S. Nakajima, T. Ka-jitani, Y. Syono, and Y. Muto, Physica C 159, 301 (1989).

IR. C. Budhani, D. O. Welch, M. Suenaga, and R. L. Sabatini,Phys. Rev. Lett. 64, 1666 (1990).K. C. Woo, K. E. Gray, R. T. Kampwirth, J. H. Kang, S. J.

Stein, R. East, and D. M. Mckay, Phys. Rev. Lett. 63, 1877(1989).S. L. Yuan et al. , J. Appl. Phys. 76, 1706 {1994).H. Raffy, S. Labdi, O. Laborde, and P. Monceau, Physica C1S4, 159 (1991).

Wan Yugui, Han Hanmin, Han Guchang, Wang Shunxi, andWang Zhihe, Mod. Phys. Lett. B 7, 601 (1993).

6G. C. Han et al. , Cryogenics 34, 613 (1994}.G. C. Han et al. , Appl. Phys. A 57, 363 (1993).Dai Yuandong, Chin. J. Low Temp. Phys. 15, 29 (1993).J. Z. Sun, K. Char, M. R. Hahn, J. H. Geballe, and A. Kapi-tulnik, Appl. Phys. Lett. 54, 663 (1989).

OA. I. Larkin, JETP Lett. 31, 219 (1980).M. Daeumling, J. M. Seuntjens, and D. C. Larbalestier, Na-ture (London) 346, 332 (1990).

~~G. C. Han et al. , J. Appl. Phys. (to be published).~3Y. Yeshurun and A. P. Malozemoff, Phys. Rev. Lett. 60, 2202

(1988);A. Houghton et al. , Phys. Rev. B 40, 6763 {1989).~4P. W. Anderson and Y. B. Kim, Rev. Mod. Phys. 36, 39

(1964).~5P. G. de Gennes, Rev. Mod. Phys. 36, 225 (1964).26V. Ambegaokar and A. Baratoff, Phys. Rev. Lett. 10, 486

(1963).V. Ambegaokar and A. Baratoff, Phys. Rev. Lett. 11, 104(1963).

Cao Xiaowen, Zhang, Tingyu, and Han Guchang, Solid StateCommun. 6S, 1091 (1988).