The purpose of this project in to investigate the structure and physical/chemical properties of iron oxides in the limit of a few atomic layers thickness. Monolayer iron oxides have been studied for decades, and are relatively well-understood. Our most recent experiments and simulations have shown that monolayer FeO grown on silver surfaces, where oxide—metal interactions are weak—show behavior similar to that expected for unsupported, two-dimensional materials, in contrast to FeO grown on metals like platinum, where relatively strong interactions dramatically alter the films’ structure and chemical behavior. While thick iron oxide films grown on metal surfaces show bulk structure (generally Fe3O4 under typical conditions), there are intermediate phases, a few layers in thickness, whose structures are not known and whose properties are not understood.
In the course of experiments on silver surfaces, we discovered, in addition to the monolayer FeO, a distinctive 3 ML phase with apparent Fe3O4 stoichiometry, but exhibiting a close-packed ion arrangement typical of FeO. X-ray diffraction measurements indicate that the phase has a unique structure, consisting of close-packed layers of tetrahedral and octahedral Fe, that is different from any bulk iron oxide. Given the weak interactions found for monolayer FeO, we postulate that this ultrathin Fe3O4 is a true 2-dimensional phase with novel and potentially useful physical and chemical properties.
Through the proposed computational work, we aim to determine: 1) the stability of the Fe3O4 structure found experimentally in comparison with hypothetical alternative structures, and especially the effect of the ion stacking sequences on the energy of formation, 2) the strength and nature of the interactions between the Fe3O4 films and the silver surfaces, and whether these interactions substantially influence the films’ properties, and 3) the predicted electronic and magnetic properties of the films, for comparison with current data and planned experiments.
We have carried out preliminary calculations for both free-standing and silver-supported Fe3O4 using the GGA+U approach as implemented in VASP. Our initial results are promising; we are able to obtain well-converged structures that reproduce experimental data well, and we observe very weak Fe3O4-Ag interactions, as anticipated. Further calculations are needed to verify these results and to thoroughly investigate the effects of spin structure, oxide-metal interface geometry, lateral strain, and ion coordination on formation energy, band structure, etc.