Interfacial magnetic coupling in oxide superconductor/ferromagnet heterostructures revealed by X-ray absorption spectroscopy


Superconductivity and ferromagnetism are usually incompatible phenomena and their mutual influence is known to be a proximity effect. Using a combination of polarisation dependent X-ray absorption spectroscopy and atomically resolved electron spectromicroscopy, the penetration of the magnetic order inside the superconductor was observed in the case of a ferromagnetic Mn oxide coupled to a Cu based high temperature superconductor. The interfacial CuO2 plane of a La1.85Sr0.15CuO4 thin film was found to develop weak-ferromagnetism due to the charge transfer of spin-polarised electrons from the La0.66Sr0.33MnO3 ferromagnet, even in the absence of direct Cu-O-Mn covalent bonding.

  • Share

The recent theoretical prediction of new intriguing physical phenomena has revived interest in superconducting/ferromagnetic heterostructures. For example, triplet superconductivity and Π-Josephson effects have been observed in mesa-junctions composed by low-Tc superconductors coupled across a ferromagnetic barrier [1]. However, while the knowledge of superconducting/ferromagnetic heterostructures based on conventional metals or alloys has considerably advanced in the last few years, the understanding of cuprate/manganite superconducting/ferromagnetic heterostructures is still patchy. For example, the weak ferromagnetism in the superconductor and the antiferromagnetic coupling between Mn and Cu spin across the YBa2Cu3O7/La0.66Ca0.33MnO3  (YBCO/LCMO) interface was explained by the formation of a Mn-3d-O-2p-Cu-3d molecular orbital [2,3]. Consequently, one should expect a different behaviour at interfaces where the Mn-O-Cu bridge is absent, and eventually a different effect on the superconductivity. For this reason, we have investigated superlattices composed of five repeat units of La0.66Sr0.33MnO3 (LSMO)m./ La1.85Sr0.15CuO4 (LSCO)n, bilayers, where the LSCO layer thickness was fixed at 23 unit cells (n=23), and the LSMO layer was varied from 50 to 15 unit cells. The superlattices were grown using reflection high energy electron diffraction (RHEED) assisted pulsed laser deposition (PLD) at the University of Twente MESA+ Institute. In these heterostructures, when the LSMO layer is on top of LSCO, the interface does not show a direct Mn-O-Cu bonding (Figure 1a). The exact stacking sequence for this type of interface was obtained by scanning transmission electron microscopy (STEM). Electron energy loss spectra (EELS) were then used to show the substantial charge transfer from the LSMO to the LSCO layer that takes place over 4-5 layers of both CuO2 and MnO2 (see Figure 1b). These measurements were performed at the Oak Ridge National Laboratory.

STEM image of the interface with LSMO on top of LSCO and the corresponding schematic arrangement of atoms and of Cu and Mn 3dz2 orbitals

Figure 1. a) STEM image of the interface with LSMO on top of LSCO and the corresponding schematic arrangement of atoms and of Cu and Mn 3dz2  orbitals (right). b) The valence of Mn as a function of the atomic layer evaluated from the O K edge spectra of the LSMO layer is close to 3.0 and progressively grows slightly above the bulk 3.33 oxidation state expected from the La/Sr ratio, indicating that for both MnO and CuO2 planes the electron occupation is higher at the interface than far from it (the red line is a guide for the eye).

To probe the magnetic properties and the orbital occupation of the interface with no Mn-O-Cu direct bonding, we have performed polarisation dependent X-ray absorption spectroscopy (XAS) at beamline ID08 (now ID32).  The X-ray magnetic circular dichroism (XMCD) signal at the Cu and Mn L2,3 edges was also observed in relatively small fields, down to 0.1 T. As shown in Figure 2a, below 1 T the Cu moments align opposite to the applied field and to Mn spins, in strict analogy with the YBCO/LCMO interface [2]. However, for higher fields, the Cu dichroism reverses sign, indicating that above 2 T the Cu and Mn moments are both parallel to the external field.

Evolution vs field of the Mn and Cu L2,3 XMCD spectra of a (LSMO)m/(LSCO)23uc multi-layer covered by a 6 nm ferromagnetic LSMO layer.

Figure 2. a) Evolution vs field of the Mn and Cu L2,3 XMCD spectra of a (LSMO)m/(LSCO)23uc multi-layer covered by a 6 nm ferromagnetic LSMO layer. b) Cu L3 XAS, acquired by interface sensitive total electron yield (TEY, top panel) and bulk sensitive fluorescence yield (FY, bottom panel); Iab (circles) is the normalised intensity for linear polarisation, parallel to the ab plane; Ic (solid lines) is the corresponding intensity for polarisation almost parallel to the c-axis. c) Calculated 3d orbital occupation for Cu and Mn near the LSCO/LSMO type B interface (see Figure 1), showing a partial filling of the Cu-dx2−y2 states (normally approximately half filled), and a small reduction, at the interfacial CuO2 plane only, of the number of electron in the Cu-d3z2−r2 states. Similarly, the Mn atoms increase the electronic occupation next to the interface, so that an overall electron-rich region is created across the interface.

To identify the role of the orbital occupation on the induced magnetism, we have measured the linear dichroism in the XAS of Cu. In Figure 2b, the very strong linear dichroism, determined by high absorption of photons with polarisation perpendicular (and concomitantly low absorption for polarisation parallel) to the c axis, confirms that LSCO at the interface retains the large bulk orbital anisotropy due to the almost exclusive x2-y2 symmetry of the empty 3d states. In particular, as highlighted by the inset of Figure 2b, the total electron yield (TEY) spectrum, dominated by the signal from the interface, is two times stronger than the one measured in the bulk-sensitive fluorescence yield (FY) mode, and shifted to lower energy, indicating the formation of a state mainly composed by Cu 3d3z2-r2 orbitals. Therefore, in our LSCO/LSMO superlattices, we have found that the same magnetic reconstruction exists, as previously found in LCMO/YBCO, but without substantial modifications of the Cu orbital occupation associated with the absence of a Cu-O-Mn bridge.

These results confirm the central role played by the Dzyaloshinskii-Moriya (DM) interaction in the weak out-of-plane magnetism of cuprates, previously studied by XMCD in bulk samples [4]. The theoretical calculations show that the DM interaction ferromagnetically couples to the out-of-plane component of Cu spins of neighbour CuO2 planes, while the strong in-plane short-range antiferromagnetic coupling still dominates even at the optimal doping. Whereas at the interface, the antiferromagnetic coupling between the ferromagnetic manganite and the first two CuO2 layers dominates, so that, in low magnetic fields, the antiparallel coupling of the manganite and cuprate out-of-plane magnetisations is propagated inside the cuprate film, far from the interface by the DM interaction (Figure 2c). This mechanism is at the origin of the reduction of superconducting temperature (Tc) in cuprate/manganite superlattices.


Principal publication and authors
G.M. De Luca (a), G. Ghiringhelli (b), C.A. Perroni (a), V. Cataudella (a), F. Chiarella (a), C. Cantoni (c), A.R. Lupini (c), N.B. Brookes (d), M. Huijben (e), G. Koster (e), G. Rijnders (e), and M. Salluzzo (a), Nature Communications 5, 5626 (2014).
(a) CNR-SPIN and Dipartimento di Fisica Università di Napoli “Federico II”, Complesso Universitario di Monte Sant’Angelo (Italy)
(b) CNR-SPIN and Dipartimento di Fisica Politecnico di Milano (Italy)
(c) Materials Science and Technology Division, Oak Ridge National Laboratory (U.S.A)
(d) ESRF
(e) Faculty of Science and Technology and MESA+ Institute for Nanotechnology, University of Twente (The Netherlands)


[1] K. Senapati, M.G. Blamire and Z.H. Barber, Nature Mater. 10, 849 (2011).
[2] J. Chakhalian, J.W. Freeland, G. Srajer, J. Strempfer, G. Khaliullin, J.C. Cezar, T. Charlton, R. Dalgliesh, C. Bernhard, G. Cristiani, H.-U. Habermeier, and B. Keimer, Nature Phys. 2, 244 (2006).
[3] J. Chakhalian, J.W. Freeland, H.-U. Habermeier, G. Cristiani, G. Khaliullin, M. van Veenendaal, B. Keimer, Science 318,1114 (2007).
[4] G.M. De Luca, G. Ghiringhelli, M. Moretti Sala, S. Di Matteo, M.W. Haverkort, H. Berger, V. Bisogni, J.C. Cezar, N.B. Brookes, and M. Salluzzo, Phys. Rev. B 82, 214504 (2010).


Top image: Interfacial magnetic coupling in oxide superconductor/ferromagnet heterostructures