X-ray spectroscopy

Our group focuses on the experimental studies of magnetic and orbital properties of multilayers, superlattices, and interfaces. We use resonant x-ray and neutron reflectometry to investigate orbital and magnetic reconstructions at the interfaces of correlated oxide heterostructures, e.g. consisting of YBa2Cu3O6+y and La1−xCaxMnO3 or LaNiO3 and LaAlO3. In particular, we apply the newly developed method of orbital reflectometry to study spatially resolved orbital polarization profiles in adjacent atomic layers of transition-metal-oxide superlattices. We collaborate closely with other groups within Keimer's department, with the Andersen's theory department, the Technology Service Group as well as with external groups within the TRR 80 collaboration.


Orbital order is known to exist in many materials and often leads to massive variations of their macroscopic properties. This offers exciting possibilities of “orbital engineering”, i.e. the ability to control orbital reconstruction at surfaces and interfaces in order to tune magnetic and transport properties of artificially manufactured heterostructures, such as metal-oxide multilayers or superlattices, on purpose. However, since the amplitude of the orbital order parameter is unknown in all but a few rare and special cases, models of the link between the atomic-scale orbital order and the macroscopic properties possess little predictive power. Orbital reflectometry is a new experimental method that allows accurate quantitative reconstruction of the depth-resolved orbital polarization profiles from polarized resonant soft x-ray reflectivity data in transition-metal-oxide multilayers with a resolution of one atomic unit cell, without resorting to model calculations. That is, it can tell within an accuracy of a few percent which d-orbitals are occupied in which atomic layer. The method is based on the straightforward application of sum rules and is immediately applicable to surfaces, interfaces, and multilayers, offering the possibility to quantitatively correlate theory and experiment on the atomic scale. It can be also readily generalized to bulk diffractometry, where it will allow facile, quantitative measurements of staggered orbital order, and thus has the potential to bring orbital physics in transition-metal oxides to a new level of quantitative accuracy. As we have demonstrated, the method is sensitive enough to resolve differences of ~3% in the occupation of Ni eg orbitals in adjacent atomic layers of a LaNiO3–LaAlO3 superlattice, in good quantitative agreement with ab-initio electronic-structure calculations. It opens up new perspectives for the synthesis of transition-metal-oxide interfaces and superlattices with designed electronic properties. Potential tuning parameters for orbital engineering include epitaxial strain from the substrate and the constituents of the superlattice, the thicknesses of individual layers, and the covalency of the chemical bonds across the interface.