Abstract
We have studied by grazing-incidence x-ray diffraction the atomic structure of an ultrathin alumina film grown on Ni(111). We show that, since there is neither registry between the film and the substrate nor induced Ni relaxations, this system appears to be a prototypical freestanding oxide layer. We have been able to unambiguously determine the three-dimensional structure of the film, which consists of a substrate/Al${}_{16}$/O${}_{24}$/Al${}_{24}$/O${}_{28}$ stacking within a (18.23 \ifmmode\times\else\texttimes\fi{} 10.53 \AA{}) $R$0\ifmmode^\circ\else\textdegree\fi{} unit cell. From the different Al coordinations (3/4/5) in the layer and from the precise determination of the Al-O interatomic distances, we conclude that the film structure presents some similarities with the \ensuremath{\eta} phase of bulk alumina, which also has a high surface/bulk ratio. The precise comparison between these two structures allows us to explain that the perfect $\sqrt{3}$ ratio between the two sides of the mesh of the film is governed by the stacking of the two central planes, combining oxygen close-packed atoms below Al atoms in tetrahedral or pyramidal positions. Moreover, Al atoms at the interface plane of the ultrathin film adopt a quasitrihedral configuration, which confirms that, in the alumina \ensuremath{\eta} phase, Al atoms with such a coordination are located near the surface of the nanocrystals. The atomic structure is also very close to the one first proposed by Kresse et al. [G. Kresse, M. Schmid, E. Napetschnig, M. Shishkin, L. K\ohler, and P. Varga, Science 308, 1440 (2005)] for alumina films on NiAl(110). This strongly suggests that this atomic model, within small variations, can be extended to ultrathin alumina film on numerous other metal substrates and may be quasi-intrinsic to a freestanding layer rather than governed by the interactions between the film and the substrate.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have