← Back

Colossal Magnetoresistance Manganites and Cobaltites

Recent results

The activity in this challenging field within the Laboratory for Neutron Scattering (LNS) is illustrated by two spectacular effects: a metal-insulator transition in (La1-yPry)0.7Ca0.3MnO3 caused by oxygen isotope substitution, and a spin-state transition in La-CoO3 induced by thermal excitations.
Fig. 1: FM and AFM diffraction peak intensities in LPCM samples (y=0.8) as a function of temperature. The insert shows magneto-optical images (taken from Ref. [13]) of the surface of the LPCM single crystal (y=0.7) providing visual evidence of the mesoscopic phase separation.
Fig. 1: FM and AFM diffraction peak intensities in LPCM samples (y=0.8) as a function of temperature. The insert shows magneto-optical images (taken from Ref. [13]) of the surface of the LPCM single crystal (y=0.7) providing visual evidence of the mesoscopic phase separation.
Fig. 2: Intensity contours visualizing the temperature renormalization of the energy spectra taken for LaCoO3.
Fig. 2: Intensity contours visualizing the temperature renormalization of the energy spectra taken for LaCoO3.


Project description

The 3d-transition metal perovskite AMO3 oxides can have metallic and insulating states and different types of the long range ordering. Depending on the interatomic to intraatomic ratio of electron interaction a particular compound can be either on itinerant or localized side of Mott-Hubbard metal-insulator (M-I) transition. This kind of magnetic oxides had been attracting general attention from early 50’s as a model system in which the theoretical concepts of magnetic exchange based on the Heisenberg Hamiltonian generalized to include exchange and crystalline anisotropies can be verified. The basic interactions relevant to M-I crossover are kinetic superexchange interactions through O-2p orbitals, which give the strength of the interatomic interaction W, and an effective intraatomic energy Ueff required to add an electron to dn manifold. The majority of single valence AMO3 perovskites are antiferromagnetic insulators, however some of them lie on the itinerant-electron side of Mott-Hubbard transition (W>Ueff) showing metallic properties. E.g. SrCrO3 (3d2) is Pauli paramagnetic metal, SrCoO3 (3d5) is ferromagnetic metal with TC=212K, SrFeO3 (3d4) is a metal with antiferromagnetic spiral spin configuration. The manganese and cobalt oxides La1-xSrxMnO3 and La0.5Sr0.5CoO3 were found to be ferromagnetic metals [1] already in 1950. Ferromagnetism in the metallic mixed valence La1-xSrx(Mn3+Mn4+)O3 led to the concept of double exchange DE (a kinetic SE involving a real charge transfer, like molecular orbital, between two different states of the same Mn atom) originally formulated by Zener [2] and later developed by Anderson and Hasegawa [3] and de Gennes [4] in a more realistic DE model for itinerant electrons in the presence of localized spins. However, the magnetoresistance phenomenon was not recognized as a very important, and these particular oxides were just an example of the DE-ferromagnetic metals.

The phenomenon of magnetoresistance (MR) itself is well known for the usual metals possessing atoms with the localized magnetic moments. The electrical conductivity is increased by ferromagnetic aligning of the localized spins due to suppression of electron-ion spin-flip scattering channel. The sudden decrease in resistivity of the ferromagnetic metals as they are cooled through the TC was explained by Mott in 1936. However, in the usual “good” metals the MR effect amounts to several percents, while the colossal magnetoresistance (CMR) effect assumes a decrease in resistance by about 100%. The CMR phenomenon has attracted general attention after discovery of the huge MR in Fe/Cr artificial superlattices at room temperatures [5] in 1988. The effect was based on the fact that the coupling between ferromagnetic layers of Fe through Cr layers can be AF at certain Cr layer thickness. By applying a large enough magnetic field the magnetization of the Fe-layers line up along the field direction providing large MR-effect. The interest to this discovery was naturally generated because of the numerous technological applications in magnetic recording and sensors (note, that today all computer disk drives make use of this technology). The re-discovery of CMR-effect in the manganese perovskite oxide La2/3Ba1/3MnO3 in 1993 by Helmolt et al. [6] immediately produced a strong resonance because the value of the effect was even larger than in the artificial superlattices, thus opening expectations for a new generation of magnetic devices. Overwhelming number of subsequent studies made manganese oxides one of the main areas of research within the field of strongly correlated electron systems. Due to CMR features of manganese oxides, the interest was also re-activated to the cobaltate oxides, which also can show DE-coupling because of mixed valence of Co. It was found that the cobaltate La0.5Sr0.5CoO3 also exhibit large MR (10%, which is considerably smaller than in manganites) at metal-metal transition at TC [7].

The majority of the manganites and cobaltates are derivatives of the perovskite like oxides AMO3 (note, that there are other manganese oxides with completely different type the Mn-lattice connectivity also showing CMR effect, e.g. pirochlores Tl2Mn2O7). The mismatch of the equilibrium (A-O), (M-O) bond lengths given by geometric tolerance factor t=(A-O)/sqrt(2)(M-O) controls the type of crystal structure. For t<1 the AMO3 structure is adjusted by cooperative rotation of MO6 octahedra, which create M-O-M bond angles. The cubic ligand field splitting (between eg and t2g configurations) and the intraatomic ferromagnetic exchange are of the same order of magnitude and their balance can favor formation of high spin state (e.g. t2g3eg1, usual for mangnates) or low-/intermediate-spin (e.g. t2g6eg0, often in cobaltates). Jahn-Teller effect removes a ground state orbital degeneracy at a localized electron configuration, providing additional channel of electron lattice interaction. The strength of the interatomic interaction W depends on the electron transfer integral, which depends on both bond angle and also on the type of 3d-orbitals overlapping with O-2p orbitals. Thus, the charge, spin and lattice are strongly coupled together because of anisotropy of 3d-electrons and mixed valence of 3d-ion. A delicate balance of these interactions, which can be achieved by particular cation substitution on both A and M sites, provokes a firework of physical phenomena of current interest in modern solid state physics, such as M-I transition, mutually connected orbital, charge and spin ordering [8], polaron formation and very large isotope effects [9-11], microscopic and macroscopic phase separation [12].

References:
  • [1] G.H. Jonker, J.H. Santen, Physica 16, 337 (1950); Physica 19, 120 (1953).
  • [2] C. Zener, Phys. Rev. 82, 403 (1951).
  • [3] P.W. Anderson, H. Hasegawa, Phys. Rev. 100, 675 (1955).
  • [4] P.G. de Gennes, Phys. Rev. 118, 141 (1960).
  • [5] M.N. Babich, et al, Phys. Rev. Lett. 61, 2472 (1988).
  • [6] R. von Helmolt, et al, Phys. Rev. Lett. 71, 2331 (1993).
  • [7] M.A. Senaris-Rodriguez, J.B. Goodenough, J. Solid State Chem. 118, 323 (1995).
  • [8] D.I.Khomskii, K.I.Kugel, Phys. Rev. B 67, 134401, (2003); D.I.Khomskii, M.Mostovoy, J. Phys. A: Math. Gen. 36, 9197 (2003).
  • [9] A. Balagurov, V. Pomjakushin, D.Sheptyakov, et al, Phys. Rev. B 64, 024420-1 (2001).
  • [10] A. Balagurov, V. Pomjakushin, D.Sheptyakov, et al, Phys. Rev. B 60, 383 (1999).
  • [11] L.Fisher, A.Kalinov, I.Voloshin, N.Babushkina, K.Kugel, D.Khomskii, Phys. Rev. B 68, 174403 (2003).
  • [12] E.Dagotto, T.Hotta, A.Moreo, Physics Reports 344, 1 (2001).
  • [13] M. Tokunaga et al., Phys. Rev. Lett. 93, 037203 (2004).

Funding: ETH internal grant, SNF, Others

Partners:
Inorganic Chemistry Division, Department of Chemistry, Moscow State University, Russia , http://www.inorg.chem.msu.ru/

Contact: vladimir.pomjakushin@psi.ch, Vladimir Pomjakushin