Elementary processes during the epitaxial growth of metal oxides: MgO/MgO(001).

  • PDF / 103,843 Bytes
  • 6 Pages / 612 x 792 pts (letter) Page_size
  • 12 Downloads / 195 Views

DOWNLOAD

REPORT


E6.7.1

Elementary processes during the epitaxial growth of metal oxides: MgO/MgO(001). Gregory Geneste1; Joseph Morillo1; Fabio Finocchi2 and Marc Hayoun3 Centre d'Élaboration de Matériaux et d'Études Structurales, UPR CNRS 8011, 29 rue Jeanne Marvig, 31055 Toulouse Cedex 4, France 2 Groupe de Physique des Solides, Universités Paris 6 - 7 and UMR CNRS 7582, Campus de Boucicaut, 140 rue de Lourmel, 75015 Paris, France 3 Laboratoire des Solides Irradiés, CEA - CNRS UMR 7642, École Polytechnique, 91128 Palaiseau Cedex, France 1

ABSTRACT Elementary processes governing the epitaxial growth of metal oxides, like adsorption and diffusion, have been studied theoretically in the prototypical case of MgO/MgO(001). Most of these processes need to be investigated in an ab initio approach since they are usually accompanied by strong modifications of the electronic structure. This study has been carried out within the density functional theory, and complemented by molecular dynamics simulations using a rigid-ion potential. The growth of the oxide passes through several oxido-reduction reactions at the surface. The adsorption and the diffusion of Mg and O are strongly different from each other, and change considerably at steps from flat terraces. All these results point out the need of realistic models for the growth of metal oxides that take into account the chemical behaviour of the involved species as well as the influence of the defects.

INTRODUCTION Metal oxides are nowadays used in many applications (microelectronics, magnetic devices, etc). They may appear as epitaxially grown thin films on various substrates, substrates, or both of them. Our ability to perform realistic simulations of the epitaxial growth of these metal oxides is a theoretical challenge with actual technological implications, since reliable models can permit a better control of their fabrication process. However, the involved time and size scales are out of reach of any first-principles methods. Presently, the simulation of growth can be addressed only through Kinetic Monte Carlo (KMC) techniques. The validity of such simulations will be a direct result of the pertinent choices made for the elementary processes taken into account and the quality of the data describing them. The understanding and quantitative knowledge of the basic mechanisms (adsorption, diffusion, etc.) that govern the epitaxial growth is therefore important for technological applications. In the case of elementary metals and semiconductors the epitaxial growth, which has been extensively studied, can be divided roughly into two families of ele mentary processes: i) those occurring at the initial stage of the deposition, when the atomic species land onto the surface, up to the point they can re-evaporate or reach a thermal equilibrium with the surface. ii) Their further thermal diffusion, adsorption and coalescence, leading to the nucleation and mesas growth. The competition between these two steps can be experimentally tuned through the deposition speed and temperature, which in turn