Enhancing Electron Correlation at a 3d Ferromagnetic Surface

By covering a surface with an ordered layer of atoms or molecules, in technical jargon called adsorbates, it is possible to create an interface with desired electronic and magnetic properties. Such nano-engineered materials may find application in fields ranging from catalysis to electronics, paving the way towards their implementation in devices and sensors with improved performance.

By employing spin-resolved momentum microscopy at the NanoESCA beamline at Elettra, we studied how adsorbates affect the properties of a model ferromagnetic surface. Namely, we characterized high-purity Fe(100) films and investigated the changes induced by a single layer of oxygen atoms arranged in the well-known Fe(100)-p(1x1)O surface structure.

The momentum microscope equipped with a 2D spin detector at the NanoESCA beamline provides direct access to the band structure of the surface under study. In practice, momentum maps of the full Brillouin Zone are acquired as a function of the photoelectron kinetic energy. The spin-resolved maps, evaluated as a weighted difference of images acquired at different scattering energies from the spin filter target crystal, are then combined in a three-dimensional dataset (Fig. 1a and 1b). The spin-resolved momentum distributions are readily obtained by performing cross-sectional cuts along the high symmetry directions (red lines in Fig. 1a). The spin-resolved electronic band structure of Fe(100)-p(1x1)O is shown in Fig. 1c.

Figure 1 of the tops-story by Janas et al.

Figure 1: a) Spin-resolved momentum map of the Fe(100)-p(1x1)O surface at the Fermi energy. The surface First Brillouin Zone is highlighted by a violet square and the high symmetry points are connected by red lines. b) Stack of momentum maps collected at different kinetic energies of the photoelectrons. c) Experimentally obtained spin-resolved band structure of Fe(100)-p(1x1)O in contrast to d) the calculated band structure, including DMFT corrections to embody electron correlations.

Our experimental findings are supported by state-of-the-art theoretical simulations (Fig. 1d) that account for how electrons influence each other in a metal. Such mutual influence is referred to as electron correlation. Theoretically, it can be expressed by the effective Coulomb parameter U and the effective exchange interaction J. In general, it is computationally costly to include electron correlation in a simulation. Only utilizing the DFT+DMFT (density functional theory + dynamical mean field theory) approach, as was done in this study, electron correlation is appropriately taken into account, which may strongly affect the computed surface electronic structure. This can be appreciated in Fig. 2, where the surface density of states for both Fe(100) and Fe(100)-p(1x1)O are shown without (Fig 2a) and with (Fig. 2b) electron correlation effects.

Figure 2 of the tops-story by Janas et al.

Figure 2: Spin-resolved density of states (DOS) of both Fe(100) and Fe(100)-p(1x1)O calculated by means of a) DFT and b) DFT including DMFT corrections. The DOSs are projected on the Fe and O surface atoms for the respective systems. c) Schematic of how adsorbates can influence electron correlation at a ferromagnetic surface and how the overall properties are ultimately affected.

The inferred impact of correlation is summarized schematically in Fig. 2c: the Fe d-bands near the Fermi energy are severely narrowed compared to the case of the sole DFT calculation, and their exchange splitting is reduced. Conversely, electronic correlations lead to a spin-dependent broadening of the oxygen-related bands at higher binding energies due to the influence of emerging satellite features that are a direct manifestation of many-electron behavior.

Our results show that the concepts developed to understand the physics and chemistry of adsorbate-metal interfaces, ubiquitous in spintronics and catalysis, need to be reconsidered in light of many-body effects. Being of utmost importance, they may affect chemisorption energy, spin-transport and magnetic order, and even play a key role in the emergence of ferromagnetism at interfaces between non-magnetic systems.

This research was conducted by the following research team:

David M. Janas1, Andrea Droghetti2, Stefano Ponzoni1, Iulia Cojocariu3,4, Matteo Jugovac3, Vitaliy Feyer3,4, Miloš M. Radonjić5, Ivan Rungger6, Liviu Chioncel7, Giovanni Zamborlini1 and Mirko Cinchetti1

1 Department of Physics, TU Dortmund University, Dortmund, Germany
2 School of Physics & CRANN, Trinity College, Dublin, Ireland
3 Elettra - Sincrotrone Trieste S.C.p.A., Trieste, Italy
4 Peter Grünberg Institute (PGI-6), Forschungszentrum Jülich GmbH, Jülich, Germany
5 Institute of Physics Belgrade, Belgrade, Serbia
6 National Physics Laboratory, Teddington, United Kingdom
7 Theoretical Physics III, University of Augsburg, Augsburg, Germany

Contact persons email:

Local contact person email:


D. M. Janas, A. Droghetti, S. Ponzoni, I. Cojocariu, M. Jugovac, V. Feyer, M. M. Radonjić, I. Rungger, L. Chioncel, G. Zamborlini and M. Cinchetti, “Enhancing Electron Correlation at a 3d Ferromagnetic Surface”, Adv. Mater. 35, 2205698 (2023); DOI: 10.1002/adma.202205698

Last Updated on Friday, 24 February 2023 10:55