Scanning-tunneling-microscopy images of Ge adsorbed on an As-covered Si(001) surface

Single Adatom Exchange in Surfactant-Mediated Epitaxial Growth

“…perpendicular to the As dimer rows, and find their activation energies to be 0.5 and 0.7 eV, respectively. For a broken-dimer geometry in which the Si atom intervenes between the As atoms, the energy is reduced by about 0.7 eV; however, there is an energy barrier of 0.7 eV for the Si atom to break the As dimer, similar to the result for a Ge adatom [4]. Since this activation energy is higher than the energy barrier of 0.3 eV for surface diffusion from the site C, it is unlikely for the broken-dimer geometry to occur kinetically.…”

“…We find the seed geometry to be more stable by 0.2 eV than for the broken-bond geometry with a Si dimer on the As dimer row. In Ge adsorption, although the Ge dimer adsorbed on the As dimer row was shown to be more stable by 0.7 eV than the seed geometry [3,4], there is a barrier of 0.7 eV per Ge atom to break the existing As dimer to form the Ge dimer on the As dimer row. Although the seed geometry was used as a starting configuration in epitaxial growth of Ge [3], we point out that this geometry results from the adatom exchange after rapid incorporation of individual adatoms, not from the dimer exchange.…”

“…In an epitaxial film, surfactant-mediated growth has been receiving considerable attention mainly because it changes the growth mode from island to layer-by-layer growth [1][2][3][4][5][6][7]. For instance, in the Ge͞Si system with a large lattice mismatch, the epitaxy of Ge exhibits layerby-layer growth up to a few monolayers, followed by island formation.…”

The Surfactant Effect in Semiconductor Thin-Film Growth

“…This study was also based on first-principles calculations of the energetics and addressed Si epitaxy on Si(100) with As acting as surfactant. In this work it was established that the exchange of a Si adatom with a sublayer As site involves an energy barrier of 0.1 eV, which is considerably lower than the energy barrier for diffusion (of order 0.5 eV) or the energy barrier for dimer exchange (of order 1.0 eV) which had been invoked as a possible mechanism in earlier studies of the same system [54,173,174,175,176,177]. This is a very interesting suggestion, but falls short of providing a complete picture of the surfactant effect.…”

“…It appears however that these two effects, that is, the strong repulsion of ad-dimers and the requirement of their presence at neighboring sites for the initiation of exchange, would be incompatible as far as growth is concerned. Both the work of Yu et al [173,174,175] and the work of Ohno [176,177] deal with mechanisms in which the basic unit involved in the exchange process is a deposited dimer, as was originally suggested by Tromp and Reuter [54].…”