Directk-Space Mapping of the Electronic Structure in an Oxide-Oxide Interface

Abstract

The interface between LaAlO(3) and SrTiO(3) hosts a two-dimensional electron system of itinerant carriers, although both oxides are band insulators. Interface ferromagnetism coexisting with superconductivity has been found and attributed to local moments. Experimentally, it has been established that Ti 3d electrons are confined to the interface. Using soft x-ray angle-resolved resonant photoelectron spectroscopy we have directly mapped the interface states in k space. Our data demonstrate a charge dichotomy. A mobile fraction contributes to Fermi surface sheets, whereas a localized portion at higher binding energies is tentatively attributed to electrons trapped by O vacancies in the SrTiO(3). While photovoltage effects in the polar LaAlO(3) layers cannot be excluded, the apparent absence of surface-related Fermi surface sheets could also be fully reconciled in a recently proposed electronic reconstruction picture where the built-in potential in the LaAlO(3) is compensated by surface O vacancies serving also as a charge reservoir.

Temperature-dependent band structure ofSrTiO3interfaces

“…23,48,59 Notably the xz (and yz) bands are more than an order of magnitude weaker than the xy bands in our calculations; and while the relative intensities of the bands observed in ARPES depend on matrix elements, the d xz/yz bands are indeed considerably weaker than the d xy bands.…”

Electronic Reconstruction at the Interface Between Band Insulating Oxides: The LaAlO3/SrTiO3 System

“…A completely overview of these approaches can be found in two recent papers, which discuss the problem of formation of a q2DES at the LAO/STO interface in quite complete fashion, but with different approaches [56,57]. Here we just argue that the most likely defects, which can effectively at the same time give a solution of the polar instability problem and can explain the formation of a conducting LAO/STO interface, are oxygen vacancies at the LAO surface, also considered in recent papers [58]. The formation energy is, according to theoretical calculations, dependent on the electric field accumulated in LAO, so that only at a critical thickness, comparable to the one experimentally observed, such defects are stable in oxidising atmosphere at the AlO 2 surface.…”

Interface superconductivity

“…In the former case, one needs to explain how magnetism may coexist with superconductivity, given that estimates of its characteristic energy scale lead to values one order of magnitude larger than the Rashba spin-splitting and four order of magnitudes larger than the superconducting gap [60,66]. In the latter case, the concentration of defects that is required to account for the number of moments would be extremely large [67]. At this stage it is not even agreed that all samples consistently exhibits magnetic signatures [68,69] and, for those who do, one needs to clarify the role played by the fabrication technique, and in particular the issue of annealing [70].…”