HAL Id: jpa-00230271
https://hal.archives-ouvertes.fr/jpa-00230271
Submitted on 1 Jan 1990
HAL
is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire
HAL, estdestinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
STRUCTURAL INVESTIGATIONS OF
POLYCRISTALLINE YBCO THIN FILMS AND CERAMICS BY X-RAY AND STEM
F. Hosseini Tehrani, J. Chazelas, R. Kormann, L. Mercandalli, R. Cabanel, G.
Garry
To cite this version:
F. Hosseini Tehrani, J. Chazelas, R. Kormann, L. Mercandalli, R. Cabanel, et al.. STRUC- TURAL INVESTIGATIONS OF POLYCRISTALLINE YBCO THIN FILMS AND CERAMICS BY X-RAY AND STEM. Journal de Physique Colloques, 1990, 51 (C1), pp.C1-1061-C1-1066.
�10.1051/jphyscol:19901165�. �jpa-00230271�
STRUCTURAL INVESTIGATIONS OF POLYCRISTALLINE YBCO THIN FILMS AND CERAMICS BY X-RAY AND STEM
F. HOSSEINI TEHRANI* , J
.
CHAZELAS' * , R. KORMANN, L. M. MERCANDALLI ,R. CABANEL and G. GARRY
Thornson-CSF, LCR, Domaine de Corbeville, F-91401 Orsay Cedex, France ' ~ a b o r a t o i r e de Genie Electrique d e Paris, Ecole Superieure
d'Electricit6, Plateau du Moulon, F-91192 Gif sur Yvette Cedex, France hornso son-CSF, DRT, RCM, 178 Boulevard Gabriel Peri, F-92240 Malakoff, France
Tne YBCO thin films were deposited on a Yttria-Stabilized Zirconia (YSZ 9%Y) (100) single crystal substrate by rf magnetron sputtering@). The target was prepared by sintering a reacted mixture of Y203. BaC03 and CuO powders followed by a high temperature annealing under an O2 atmosphere(7)at 5 0 ° C during 4 hours. The films were de- posited up to a thickness of Ipm on a non intentionally heated substrate (below 10°C). The films were amorphous and so needed a post-thermal treatment under an ambient.
Surface crystallization has been studied under ultra-high vacuum in a 100 KeV scanning transmission electron mi- croscope ( STEM VG HBSOIA ). Ex-situ annealing has been made in a classical furnace at atmospheric pressure with controlled gas composition, the rise of temperature was canied out under He gas with or without O2 while the cooling was performed under pure 02.Their lattice parameters and preferred orientation were measured with a Seifert MZ-IV X-ray goniometer in vertical position, used in a symmetrical Q-2Q geometry with CuKa radiation. The metal compo- sition of the samples was obtained by an electron probe microanalyses (EPMA) and the electrical resistance was measured by a conventional four-probe method. Roughness measurments were performed with a profilometer
.
2lRESULTS
a: High vacuum annealing
The sample was annealed up to 850°C, using the preparation chamber ( I O - ~ tom) of the microscope. A step by step procedure was used, consisting in a series of annealing cycles at higher and higher temperatures. After each step (room temperature,200°C,4000C,5400C,7200C,8500C), the variation of the surface sample composition was analysed by Auger scanning microscopy, in conjunction with the variations of the surface topography, using the reflexion mode ability of the STEM. The samples reached the desired temperature in 1 hour, to which they were maintained for 5 to 15 mn, and then cooled in l hour.
The surface transformation is clearly visible on Figs.l,2 and 3: the first micrograph was taken on the edge of a hole in the amorphous film, showing the smoothness of the film and the substrate surface. It is only above 5M°C that some blistering appears, as shown on second micrograph although the Auger analyses showed no significant surface differencies with the surroundings.
Finally, around 800°C, the surface becomes rough, as shown on the third micrograph, where a "liquid phase"
seems to appear. This temperature is similar to that observed by Ullman suggesting the following reactions:
Rtsurnt: Des Ctudes de couches minces YBCO men& parallClement A celles de ~Cramiques, nous poussent k croire que I'orientation prkfkrentielle des couches minces dCpend du taux d'Oz prksent dans I'aunosphkre de recuit. Nous dkcrivons la dkmarche des ktudes, et proposons un mkanisme de nuclkation.
Abstract: A preliminary comparative study of YBCO thin films and ceramics is described, suggesting a dependence of the film orientation on the oxygen gas pressure during annealing. A nucleation mechanism is also proposed.
The electrical transport properties of a polycristalline superconductor depend strongly on the microstructure fea- tures (texture, grain boundaries, microcrackings ...), so the knowledge of the structural parameters is required for a re- producible procedure to obtain a good superconducting material. With this aim, we have observed the microstructure of various materials with various electrical properties: thin films and ceramics.
Whilst the best YBCO films are now obtained on SrTi03 single crystals (with critical current density Jc,77~K=106 A/cm2, and Tc=900K)('), on the other hand, the best bulk materials have never exhibited a Jc value higher than 104 A/cm2 (2). Therefore, it is important to investigate the structural difference between bulk materials and thin films to un- derstand the relationship between the structural and the superconducting properties.
Up to now, superconducting films have been prepared as-grown by an appropriate substrate temperature, or more classicaly, by annealing at high temperature after deposition on a non heated substrate. However, as some groups have reported, the films have different orientations depending on the substrates, the preparation conditions(3.4) and the 0, partial pressure (5). Factors controlling the orientations of the superconducting crystals have not yet been clarified.
This paper reports the effects of the O2 partial pressure during the post-annealing treatment of thin films, and pro- poses a model for the growth and crystallization of the superconducting phase. A comparison is also made between two ceramics with different critical current densities.
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19901165
COLLOQUE DE PHYSIQUE
Y Baz Cu307-, + BaCuOz
+
CuO <=> liquid (eutectic "l") BaCu02+
C u 0 <=> liquid (eutectic "2",8 16OC).Relative composition variation has been determined through comparison of the variation of the Cu(LMM)/O Ba(MNN)/O(KLL) and Cu(LMM)/Ba(MNN) Auger signal intensity ratios with temperature (Fig.4). The decr the copper signal sug ests either a migration of the cations into or out of the film, or the surface segregation c phases by at least 30% (corresponding to the Auger electron escape depth), screening Cu atoms. This is cor with the increase of the BafO peak ratio. Moreover, it is known@)that, under 0 2 atmosphere, the O2 outdiffus start at 2W°C and becomes maximum at 500°C; this phenomenon also occurs in our sample but at temperature,probably activated by the high vacuum condition.
A comparison was made to a sample, annealed in the classical way@) (without O2 up to 880°C, plateau o.
and cooling under O2 ), and exhibiting superconductive properties (TC=9O0k). The surface aspects (Fig.5) (
show the differences between the two annealing methods and the results of ~ u ~ e r analyses are given in Fig.4 cal b : Heliwn and Oxygen anneatin(:
The samples A and B (YBaCuONSZ) deposited simultaneously, exhibited a 1/2.04/3.12 composition~measu EPMA. The annealing parameters and the results of the different analyses used for this study are summarized in 1. For sample A, the annealing was carried out under pure He gas, whereas sample B was heated under a partia sure of 02. The roughness measurements revealed that samples annealed under an 0 2 partial pressure are sm than those annealed under He. On the other hand, the transition temperature is much higher for the He heated sa:
Figs.6 and 7 display the results of X-ray diffraction analysis, for A and B, respectively. The samples are bo ented, preferentially with the c axis normal to the subsrate surface, as shown by the rocking curve, but for the fi the majority of the grains are epitaxially grown, whereas the c axis of the sample B makes an angle of 32' with r to the substrate c axis.
U1 -
BULK STUDYCeramic samples of chemical com~osition Y1Ba2Cu307-x were repared by conventional cold pressing an (16f
casting methods (sintering temperature in the range 900-950°C) a ceramic Precursor provided by R Poulenc. The cold pressed sample was sintered at 950°C during 16h, whereas the tape casted one (of 50pm thic was sintered at 91S°C/16h. The cooling of both undertook a plateau at 500°C/4h.
The key parameter which differentiates these samples is the critical current density: 100AJcm2 and 1000A/c spectively
.
Both samples were studied by STEM and X-ray microanalyses (Figs.8 and 9). The samples were prepared b:
followed by an Ar ion beam milling (4keV) at liquid nitrogen temperature for the cold pressed, whereas the tape sample needed only the ion milling in the same conditions. We have observed the classical twins(ll), only a fraction of the mamx being in the untwinned form. For the cold pressed sample, where the grain size was 100pm, an amorphous phase was sometimes observed on electron diffraction pattern of the grain boundaries, , the EDX analyses showed mostly a Ba rich phase, but with a variable composition. The grain boundaries of tk casted sample are in contrast very clean. The grain size is about 10pm. A secondary phase is observed, but we some grains without twinning. One could deduce that these grains are tetragonal, as the twins appear with straj companying the tetragonal to orthorhombic transformation. However, some recent (l2J3) works have shown th der proper conditions, it was possible to have orthorhombic grains without twinning.
We have observed that, when annealing an amorphous YBCO thin film under an inen gas(He), grains gro\
taxially, whereas the effect of an O2 partial pressure seems to be the rotation of the c axis until being in the plane substrate in the case of a pure 02These experimental results and others under pure 02(6), suggest the following r ation and growth mechanism (Fig.10).
The presence of epitaxial grains suggest the subsuate interface to be the nucleation site. Reminding O2 outdiff is activated under pure He, the 0 2 diffusion may occur at the YBCOIYSZ interface, from the rich phase (YSZ) poor phase (YBCO). Due to this interdiffusion an O2 concentration gradient arises, perpendicular to the inte producing planes composed of the same kind of atoms, the one with the more stable oxidated state. In order to rnize the system energy, a seed with c axis normal to substrate therefore arises.
On the other hand, in the presence of oxygen in the annealing atmposphere, the different atoms are oxidizec no marked preference, but mainly at the sample surface, where BaO-CuO-Y-
...
planes build up. Moreover, the, sion and concentration gradient are of opposite sign as compared to the preceding case, as the oxygen adsol mechanism is dominant below 500°C(9). Thus, the direction of the preferential growth is the same as in ceramic perpendicular to the substrate c axis for energetical reasons.Throughout a preliminary study of the annealing parameters of thin fims and ceramics, we suggest a nuclc mechanism. The validity of this explanation could be corroborated by using TEM observations on samples wi
c--n A r n n t n n t lerren xxr;thr\nlr 0.) hntt o n n e ~ l e r l ; n tr~Ann.r 0. n Q v 4 - l nrercr,re n t m n r n h e r ~ r anrl a l m f i l m r wit
ACKNOWLEDGMENTS
R. A. Kreisler, and Dr. D. Dubreuil are gratefully acknowledged for support and encouragement. The authors would like to express their thanks to Dr. R. Bisaro for fruitful comments on X-ray results. Special thanks go to Dr.
A: Friederichor helpful discussions.
l ) D . M . w a n g , T . V e n k a t e s a n , C . C . C h a n g , L . N a z ~ , 1702 (1989).
2)S.Jin,T.H.Tiefel,R.C.Sherwood,R.B.van Dover,M.E.Davis: Phys.Rev.B32,7850(1988)
3)P.Chaudhari,R.H.Koch,R.B.Laibowitz,T.R.McGuire,R.J.Gambino: Phys.Rev.Lett.58.2684(1987) 4)D.K.Lathrop,S.E.Russek,R.A.Buh1man: Appl.Phys.Lett~,l554(1987)
5~Y.Terashima.M.Sagoi,K.Kubo,Y.Mizushima: Jap.Jour.of Appl.Phys28(4),L653,(1989)
6)D.Dubreuil,G.Gany,Y.Lernaitre,L.Rogier,D.Dieme ard: Jour.of Less Common Metals, m,303,(1989)
7)R.Kormann.J.P.Ganne: Proceedings of 'tJournkes d'iudes sur ies ckramiques supnconductrices d haute TCf' Uni- versitC de Caen (septembre 1988) p108.
8)J.E.Ullman,R.W.McCallum,J.D.Verhoeven: J.Mater.Res.,u,752(1989) 9)S.I.Shah: Appl.Phys.Lett.53(7),612,(1988)
lO)J.P.Ganne,R.Korman,M.Labeyries,F.Lainee,B.Lioret: Proceedings Stanford September (1989) 1 l)J.Y.Laval,M.H.Berger,C.Delamarre,H.Zeng,A.Dubon: Revue Phys.AppI.a,489(1989) 12)Z.Z.Wang,N.P.Ong,J.T.McGinn: J.Appl.Physm,2794(1989)
13)D.L.Kaiser,F.W.Gayle,R.S.Roth,L.J.Swartzendruber: J.Mater.Res.,14(4),745((1989)
Fig. l Before annealing
Above 500°C
YBCO
Fig. 2 Blister t r a n s f o r p a t i o n
0 . 5 ~ 5 P
H H
-the surface becomes rough
Fig.3 around 800°C
Fig.5: Comparison with a classical anneal
*
Laboratoire de Genie Electrique de Paris, Ecole SupCrieure d'Elecnicit6, Plateau du Moulon, 91 192 GIF SUR YVETTE CEDEX**Thornson-CSF, DRT, RCM, 178 Boulevard Gabriel Peri 92240 MALAKOFF
COLLOQUE DE PHYSIQUE
Table 1 : Summary of deposition aan- md differenr characterisation
-1
.-
'*
I I I , I I I , , I I I I , I I , , I , , , , , , I , , , , I , ,
0 200 400 600 800
T E M P E R A T U R E (
C
)Fig4
: Variation of the Cu/O, Ba/O and Cu/Ba Auger Intensity Ratio with the annealing temperature.--
• ( 4 Y l B d C o 3 0 . b(A)
I
$81 (4) 3,89 (0) 11,- (6) 3 s (1) -
p.t c-)
0,ZS 0,s
0,15 0,lS
Resistivity versus temperature
YZBc(ClbOy c (A)
n,a
(6) n,24 (2)Tc (AT to-,,%)
90
(121
70 (16,7) W
A
B
--
Frabriqamb. T CC)(l.Da)%(PC16 Hdi\lm(lrhn)
104 2,5%G
(c) rocking curve (001)
1
5.0 10.0 1 5 . 0 20 . o 25.0 30.0 35.
(c) rocking curve (001) Fig.7: 0 - 2 0
ofsample
A@)with substrate intensity max; (b)with rocking the substrate
(hkl)= Y lBa2Cu30y <hkl>=Y2Ba4Cu80x
Unknown phases: (* 21.14; 50.15),(**15.52;28.91;31.41;48.84;64.45)
With 0 2
c o n c e n t r a t l o n
p gradient o f -02-
H e a t i n g
kzEms-
T" < 500'~ llrnltlng f a c t o r.
surface oxidalWithout 0 2
C VBCO
-
c o n c e n t r a t i o n n t o f '02"N u c l e a t i o n
H e a t i n g
l i m i t i n g f a c t o r = i n t e r f a c e oxygen d i f f u s ~ o n
Fig.10 nucleation and growth mechanism
Cl-1066 COLLOQUE DE PHYSlQUE
Fig.9
@ @ @
the tape casted sample10
nm
H
amorphous phase
LL&L,
Figs.8 The cold pressed sample STEM and X-ray microanalyses
