Oxide heterostructures and superlattices

Modern microelectronics relies to a large degree on the properties of interfaces between two different semiconductors. In a transistor, for instance, the electrical conductivity at the interface can be controlled by an external voltage. Interfaces between transition-metal oxides (TMOs) can exhibit a plethora of phenomena not observed in ordinary semiconductor heterostructures such as superconductivity or ferromagnetism. Even the interface between the relatively simple materials SrTiO₃ and LaAlO₃, which are both band insulators, exhibits a rich phenomenology: "Electronic reconstruction" has been observed, leading to a metallicity at the interface, and two-dimensional superconductivity at low temperatures. Lateral structuring techniques are now being explored vis-à-vis electronic device applications.

Further degrees of freedom can play a role in the interfacial physics, such as magnetism and orbital physics. The emphasis of this project is on the fundamental understanding of these latter phenomena, which may open a path towards a new generation of electronic devices. This project is an excellent example of how the synergetic effects due to the variety of spectroscopic and theoretical methods employed in our department, in combination with support from Service Groups at our institute, can be used to the advantage of our investigations.

A major emphasis is being put on the synthesis of manganate and cuprate thin films and heterostructures in the Technology Service Group.

Early optical ellipsometry work indicated an interfacial reconstruction and a magnetism-induced suppression of metallicity in superlattices comprised of superconducting YBa₂Cu₃O_6+x and ferromagnetic La_0.7Ca_0.3MnO₃ [11]. The magnetization profile was probed directly using neutron reflectivity (Fig. 1) [10]. In combination with x-ray circular dichroism measurements, we showed that the Cu ions are ferromagnetically polarized in a 20 Å thick interfacial layer (Fig. 2) [9].

Theoretical calculations of the Cu-Mn exchange interaction across the interface show that the bulk occupation of Cu and Mn d-orbitals cannot be maintained at the interfaces and imply that an orbital reconstruction occurs: The Cu valence electrons reside predominantly in the 3d_z²-r² orbitals, perpendicular to the interface, in contrast to the bulk (Fig. 3, 4). This was experimentally confirmed by x-ray linear dichroism measurements.

Predictions of orbital order and superconductivity at the interface between LaNiO₃ and a band insulator like LaAlO₃ by our Theory group motivated an experimental effort to grow (Fig. 5) thin films and superlattices comprised of these materials in the Technology Service Group. Currently, we are investigating these samples in the Neutron & x-ray reflectometry group, the Optical spectroscopy group and the Inelastic photon scattering group.

Further theoretical efforts are related to the behavior of correlated electrons at LaVO₃ - SrTiO₃ interfaces and to "orbital engineering" at LaNiO₃ - LaAlO₃ interfaces [4, 6]. Finally, we also closely collaborate with external groups in the endeavor to improve our understanding of the superconductor - ferromagnet interface found in La_1-yCa_yMnO₃ - YBa₂Cu₃O_6+x heterostructures [3], as well as the paramagnetic-metal - antiferromagnetic-insulator interface found in CaRuO₃ - CaMnO₃ heterostructures.