Evolution of Ge islands on Si(001) during annealing

Abstract

The evolution of the shape and size distributions of Ge islands on Si(001) during annealing after deposition has been studied at different temperatures and effective coverages. The initial distributions of square-based pyramids, elongated “hut” structures, faceted “dome-shaped” islands, and much larger “superdomes” depends on the deposition conditions. During annealing after deposition, the islands coarsen over a limited range of times and temperatures. Those pyramidal-shaped islands that grow transform to faceted, dome-shaped islands as they become larger. Initially dome-shaped islands that dissolve transform to a pyramidal shape as they become smaller during the process of dissolving. Outside of this coarsening regime, the islands can achieve a relatively stable, steady-state configuration, especially at lower temperatures. At higher temperatures, intermixing of Si into the Ge islands dominates, decreasing the strain energy and allowing larger islands to form. At lower and intermediate temperatures, the initial wetting layer is metastable, and some Ge transfers to the islands during the early stages of annealing.

Controlling the quantum dot nucleation site

“…These findings are in agreement with the theoretical predictions of Shchukin et al [45]. It has been also verified [46] that the island nucleation and evolution is independent from the growth method. Capellini et al [47 Á/50] and Goryll et al [51,52] also report on high and low pressure CVD growth of Ge/Si(001) islands, confirming these results, and extending the analysis to dislocated islands.…”

“…The overall process can be qualitatively described as follows: the islands grow vertically up to a critical height, which has been estimated to be about 48 nm [68] after which the strain energy stored inside the islands can be partially relieved by introducing dislocations, or by a morphological transition of the island which progressively becomes more rounded in shape. Concerning the substrate erosion around the island, a similar effect was previously reported by Kamins et al [46] on Ge/Si(100); however, they could not understand clearly its origin due to the oxidation of their samples (the measurements were performed by AFM ex situ, in air) which prevented a clear imaging of the trench.…”

“…At high island density Ross et al [54,55] found that smaller islands shrink during annealing due to evaporation. Kamins et al [46] also observed a ripening effect which enlarges some of the islands at the expenses of the smaller ones; however, at sufficiently low temperatures the distribution is stable for the two equilibrium shapes: domes and pyramids. Therefore, Kamins et al state that the coarsening behavior is not consistent with standard Ostwald ripening models [56].…”

“…[46], in order to verify if pyramids and domes are really the most stable structures, or if they are subject to Ostwald ripening, i.e. if the particle size distribution coarsens, driven by the Gibbs Á/Thomson effect [53].…”

Growth and characterization of InAs columnar quantum dots on GaAs substrate

“…[28][29][30] The PL peaks related with the QW slightly blueshift, indicating that the thickness of the InAs layer in the QW partly reduces during the GI to provide material for the CQD growth. 29,30 However, this transfer of InAs material from the QW may only play a minor role in achieving the larger PL peak redshift for the CQDs. The volume increase of the CQDs can be mainly ascribed to material exchange among different CQDs.…”