A MEIS Study on Ge Eppitaxial Growth on Si(001) with dynamically supplied Atomic Hydrogen

  • Ha, Yong-Ho (Department of Chemistry, KAIST) ;
  • Kahng, Se-Jong (Department of Physics, Seoul National University) ;
  • Kim, Se-Hun (Department of Chemistry, KAIST) ;
  • Kuk, Young (Department of Physics, Seoul National University) ;
  • Kim, Hyung-Kyung (Surface analysis group, KRISS) ;
  • Moon, Dae-Won (Surface analysis group, KRISS)
  • Published : 1998.02.01

Abstract

It is a diffcult and challenging pproblem to control the growth of eppitaxial films. Heteroeppitaxy is esppecially idfficult because of the lattice mismatch between sub-strate and depposited layers. This mismatch leads usually to a three dimensional(3D) island growth. But the use of surfactants such as As, Sb, and Bi can be beneficial in obtaining high quality heteroeppitaxial films. In this study medium energy ion scattering sppectroscoppy(MEIS) was used in order to reveal the growth mode of Ge on Si(001) and the strain of depposited film without and with dynamically supplied atomic hydrogen at the growth thempperature of 35$0^{\circ}C$. It was ppossible to control the growth mode from layer-by-layer followed by 3D island to layer-by-layer by controlling the hydrogen flux. In the absent of hydro-gen the film grows in the layer-by-layer mode within the critical thickness(about 3ML) and the 3D island formation is followed(Fig1). The 3D island formation is suppressed by introducing hydrogen resulting in layer-by-layer growth beyond the critical thickness(Fig2) We measured angular shift of blocking dipp in order to obtain the structural information on the thin films. In the ppressence of atomic hydrogen the blocking 야 is shifted toward higher scattering angle about 1。. That means the film is distorted tetragonally and strained therefore(Fig4) In other case the shift of blocking dipp at 3ML is almost same as pprevious case. But above the critical thickness the pposition of blocking dipp is similar to that of Si bulk(Fig3). It means the films is relaxed from the first layer. There is 4.2% lattice mismatch between Ge and Si. That mismatch results in about 2。 shift of blocking dipp. We measured about 1。 shift. This fact could be due to the intermixing of Ge and Si. This expperimental results are consistent with Vegard's law which says that the lattice constant of alloys is linear combination of the lattic constants of the ppure materials.

Keywords