Nanometric artificial structuration of semiconductor surfaces for crystalline growth

The coupling of standard self-organization methods with surface artificial nanostructuring has recently emerged as a promising technique in semiconductor materials to control simultaneously the size distribution, the density and the position of epitaxial nanostructures. Some physical aspects of the morphology and elastic strain engineering are reviewed in this article. The emphasis is on the effects of capillarity, growth rate anisotropy, strain relaxation and entropy of mixing for alloys. The interplay among these driving forces is first illustrated by III-V compound semiconductor growth on lithographically patterned surfaces, then by germanium growth on implanted substrates and nanopatterned templates obtained by chemical etching of buried strain dislocation networks. (C) 2004 Academie des sciences.