Magnetic ordered phase in La 0.6 Sr 0.4 MnO 3 ferromagnets L. B. Steren,* M. Sirena, and J. Guimpel Centro Ato ´mico Bariloche and Instituto Balseiro, Comisio ´n Nacional de Energı ´a Ato ´mica, 8400 S. C. de Bariloche, R. N., Argentina Received 8 July 2001; published 20 February 2002 The ferromagnetic phase of La 0.6 Sr 0.4 MnO 3 thin films has been investigated through measurements of magnetization loops at different temperatures, zero-field-cooled and field-cooled magnetization curves obtained under different magnetic fields. We have found that the main sources of the ‘‘bulk’’coercivity in manganite films are the film/substrate interface and film surface. The temperature dependence of the coercivity is de- scribed by a ‘‘strong domain-wall pinning’’ model, independently of the thickness and substrate. The magnetization-vs-temperature curves, measured under different magnetic fields (10 OeH2.5 kOe), have been explained in terms of the magnetic hysteresis of the films. We find no evidence of glass states or of the existence of single-domain clusters, as suggested by other authors. DOI: 10.1103/PhysRevB.65.094431 PACS numbers: 75.60.-d, 75.70.-i I. INTRODUCTION In recent years, manganite oxides have attracted a lot of attention due to their potential application as magnetoresis- tive sensors. A 1 -x B x MnO 3 ( A =La, B =Sr,Ba,Ca) com- pounds exhibit a wide variety of magnetic and electric trans- port properties, depending on the concentration x. 1 In the region 0.2x 0.5 these compounds present ferromagnetic order and an insulator-metal transition near the Curie tem- perature, while the undoped parent compound LaMnO 3 is an antiferromagnetic insulator. Zenner 2 proposed the existence of a double-exchange interaction between Mn 3 + and Mn 4 + to explain both the ferromagnetic order and the metallic character of the doped compounds. Many questions have been opened about the interplay of this ferromagnetic term with the antiferromagnetic superexchange in manganite com- pounds. de Gennes 3 proposed the existence of a frustrated phase in an intermediate-concentration range, based on the competition between both interactions. The important mag- netohistory effect observed in the ferromagnetic manganites has given rise to a wide variety of interpretations for the magnetic phases of these compounds. 4,5 Ju and Sohn discuss in Ref. 4 zero-field-cooled ZFCand field-cooled FCmag- netization curves and the thermal dependence of coercitivity in terms of spin freezing and magnetic inhomogeneity. These authors studied the variation of the magnetization curves with oxygen content of different bulk and film samples of La-Ca-Mn-O and La-Ba-Mn-O, and interpreted their results in the frame of a ‘‘spin-clustered’’ system. As these com- pounds are highly disordered, with Mn 3 + -Mn 4 + pairs dis- tributed randomly, the idea of clusters embedded in a non- magnetic matrix has been introduced for the explanation of some experimental results. However, the temperature depen- dence of the coercive field does not follow the expected law for blocked cluster systems. Moreover, the blocking tempera- ture deduced from the fit does not correspond to the maxi- mum in the ZFC curves. Li and coworkers 5 introduced the concept of a ‘‘cluster-glass’’ phase in La 0.7 Sr 0.3 Mn 0.7 Co 0.3 O 3 . In spite of the fact that for Co doping the interpretation of the experimental data is complicated due to the coexistence of Co in different electronic states, the features observed in magnetization measurements and attributed to a field- induced transition from a cluster-glass to a ferromagnetic phase could not be taken as conclusive on this subject. It is important to remark that irreversibilities between zero-field- cooled and field-cooled magnetization curves have been ob- served in both ferromagnetic bulk 5,6 and film samples, 4 so they cannot be only assigned to substrate influence on the magnetic properties. We believe that without performing a complete experi- mental study of the magnetism of the samples it is rather difficult to identify complex magnetic phases, like spin or cluster glasses. In this work, the field and temperature dependences of the magnetization of magnetoresistive La 0.6 Sr 0.4 MnO 3 thin films have been studied for different magnetic histories in order to understand the nature of the magnetic order in these compounds. Bulk La 1 -x B x MnO 3 (0.2x 0.5; B:Sr,Ba,Ca) is a relatively soft material H c 10 Oe Ref. 7. Due to the fact that ‘‘low-field’’magnetic measurements are usually performed in fields larger than 10 Oe, small magnetohistory effects have been reported for bulk compounds. However, in thin films, the coercive field is larger 8 and a rich variety of behaviors is observed, depending on the applied field and temperature. II. EXPERIMENTAL DETAILS The films were grown by dc magnetron sputtering from a stoichiometric ceramic target of nominal composition La 0.6 Sr 0.4 MnO 3 LSMO. Films with thickness t, ranging from 5 nm to 500 nm, were grown both on 100MgO and 100SrTiO 3 single-crystalline substrates. The growth tem- perature was 660° C. After deposition the film was cooled down slowly to room temperature in a 100 Torr O 2 partial pressure. The film composition was measured by energy dispersive x-ray analysis EDAX. The results show that the composi- tion of the films is that of the target within 10% and that it is homogeneous along the sample. The crystalline structure of the samples was characterized by x-ray diffraction spectros- copy. X-ray diffraction patterns indicate that the films are strongly textured, with the 001pseudocubic axis in the film normal direction. 9 The lattice-parameter-thickness depen- dence shows evidence of strains, for SrTiO 3 STOsub- PHYSICAL REVIEW B, VOLUME 65, 094431 0163-1829/2002/659/0944316/$20.00 ©2002 The American Physical Society 65 094431-1