Structural relaxation effects on interface and transport properties of Fe/MgO(001) tunnel junctions
Xiaobing Feng (1), O. Bengone (1), M. Alouani (1), S. Leb\'egue (2),, I. Rungger (3), S. Sanvito (3) ((1) Institut de Physique et Chemie des, Mat\'eriaux de Strasbourg, Strasbourg, France (2) Laboratoire de, Cristallographie et de Mod\'elisation des Mat\'eriaux Min\'eraux et

TL;DR
This study investigates how different computational relaxations of Fe/MgO(001) interfaces affect their electronic and transport properties, revealing significant sensitivity that impacts the accuracy of theoretical predictions versus experiments.
Contribution
It demonstrates that the choice of exchange-correlation functional in DFT significantly influences interface structure and transport calculations in Fe/MgO tunnel junctions.
Findings
Conductance varies by an order of magnitude depending on the relaxation method.
Interface charge transfer is sensitive to the exchange-correlation functional used.
Sensitivity of electronic current may explain discrepancies between theory and experiment.
Abstract
The interface structure of Fe/MgO(100) magnetic tunnel junctions predicted by density functional theory (DFT) depends significantly on the choice of exchange and correlation functional. Bader analysis reveals that structures obtained by relaxing the cell with the local spin-density approximation (LSDA) display a different charge transfer than those relaxed with the generalized gradient approximation (GGA). As a consequence, the electronic transport is found to be extremely sensitive to the interface structure. In particular, the conductance for the LSDA-relaxed geometry is about one order of magnitude smaller than that of the GGA-relaxed one. The high sensitivity of the electronic current to the details of the interface might explain the discrepancy between the experimental and calculated values of magnetoresistance.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
