Secondary-phase orientation probed by EPR: Y 2 BaCuO 5 in YBa 2 Cu 3 O 7 2x N. Pellerin, P. Simon, P. Odier, F. J. Gotor, N. J. Poirot, and L. Vutien Centre de Recherches sur la Physique des Hautes Tempe ´ratures, CNRS, 45071 Orle ´ans Cedex 2, France L. Durand Commissariat a ` l’Energie Atomique, Centre d’Etudes et de Recherches sur les Mate ´riaux, De ´partement de Technologie des Mate ´riaux, Section de Recherches de Me ´tallurgie Physique, CEN de Saclay, 91191 Gif sur Yvette Cedex, France I. Monot Cristallographie et Sciences des Mate ´riaux—ISMRA, 14050 Caen Cedex, France ~Received 12 June 1996! Textured YBa 2 Cu 3 O 7 2x samples have been studied by electronic paramagnetic resonance. EPR spectros- copy allows us to analyze the Y 2 BaCuO 5 second phase present as inclusions in the YBa 2 Cu 3 O 6 textured matrix. All samples were deoxygenated in order to limit the skin effect due to conductivity. This is applied to samples textured according to different techniques. Preferential orientation of these particles according to the 123 matrix can be observed. For instance, if texturing under a magnetic field, the particular Y 2 BaCuO 5 orientation is observed of the b 211 axis ~b 57.132 Å! parallel to c 123 and to the magnetic field direction. By appropriate calculations from initial spectra, the fraction of preferred orientation of these Y 2 BaCuO 5 grains can be quantified. This orientation comprises 1662% of the Y 2 BaCuO 5 weight for texturing under a magnetic field, presently one of the highest values obtained. It involves the longest size of Y 2 BaCuO 5 elongated grains lying in YBa 2 Cu 3 O 6 planes, facilitating the texturation process. Texturing on Y 2 O 3 by the modified melt textured growth method leads to a preferential orientation of Y 2 BaCuO 5 grains too, but in another direction, i.e., b 211 perpendicular to c 123 . There is no Y 2 BaCuO 5 preferential orientation if texturing with a thermal gradient or on a monocrystalline MgO substrate. This type of analysis can be applied easily to other systems ~nanocomposites, dispersoids, etc.!, providing that the signal of the secondary phase can be clearly separated from the signal of the primary one. @S0163-1829~97!06001-3# I. INTRODUCTION High-T c cuprates are EPR silent above and below T c . 1 The presence of bidimensional antiferromagnetic fluctuations has been proposed 2,3 to explain this, but no definite conclu- sions are yet possible. The second phase could be then at- tractively studied by this technique. CuO which is often present as an impurity in textured YBa 2 Cu 3 O 7 2x ~123! gives no visible EPR signal in X band, or very broadband. 1 The EPR signal of the BaCuO 2 1x phase, also present in small quantity, strongly depends on oxygen stoichiometry. 4 For small x values it consists of a very broad line, while when oxidized the line shape is closer to that of Y 2 BaCuO 5 ~211!. However, due to its more marked rhombic character, it is definitely different from that of 211. 5 Moreover, its smaller intensity than that of 211 allows one to neglect its contribu- tion in 123 based compounds. Then EPR allows one to study the 211 secondary phase trapped in the 123 textured matrix. 6 This spectroscopy gives access to bulk information on 211 phase, contrary to TEM, often used to study 123/211 inter- faces. Various methods allow one to elaborate bulk high-T c su- perconducting materials. 7,8 Melt texturing is one of the most promising for 123. 9 Its peritectic decomposition into the green phase 211 plus a liquid, followed by a slow cooling, allows directional growth of 123 domains. As a result of kinetic limitations, the peritectic recombination of 211 into 123 is not complete which causes 211 inclusions to remain trapped in the texture. The presence of 211 grains in the 123 matrix involves various defects which increase the flux pin- ning. However, the exact pinning mechanisms are still a mat- ter of debates. Because 211 is a witness of the peritectic decomposition and recrystallization of 123, it is a key factor for both crystallization process understanding and critical current enhancement. Its chemical and physical characteriza- tion is therefore an essential step in the development of these materials. Electron microprobe, 10 SEM, and TEM are now systematically used. Here we have exploited EPR as a way to probe the orientation of 211 inclusions with respect to the matrix. It is a rare example of such a spectroscopy applied to a second-phase analysis. The 211 phase is insulating and has an EPR signal due to the presence of magnetically independent Cu 21 ions with 1/2 spin. The only determination of the principal axes of the g tensor of the Cu 21 ion in the 211 phase has been made by Kobayashi et al., 5 on single crystals. They have obtained the values g x 52.050, g y 52.094, and g i 5g z 52.222, corre- sponding to a local rhombic symmetry. Because of a very low splitting between g x and g y , the mean value is noted g ’ . Moreover, the principal ( X , Y , Z ) g -tensor axes are respec- tively parallel to the crystallographic ~a,c,b! axis where a 55.658 Å, b 57.132 Å, and c 512.181 Å. This result is consistent with our x-ray pole figure on 211. 11,12 The a crys- tallographic axis lies parallel to the longest dimension of the elongated particles with a typical size of less than 20 mm. In previous papers, we have reported qualitative EPR re- PHYSICAL REVIEW B 1 JANUARY 1997-II VOLUME 55, NUMBER 2 55 0163-1829/97/55~2!/1262~7!/$10.00 1262 © 1997 The American Physical Society