Competing growth mechanisms of Ge/Si(001) coherent clusters

Abstract

The growth of Ge three-dimensional coherent clusters on Si͑001͒ during gas source molecular-beam epitaxy and post-deposition anneals has been investigated using in situ elevated-temperature scanning tunneling microscopy. By monitoring the growth of individual so-called ''hut'' clusters, this technique allowed us to separate various factors that may affect the final size distribution of entire cluster ensembles. It has been found that during the course of epitaxy the hut clusters grow by nucleation and growth of deposited material on the cluster facets; however, the low growth rate ͑rϰt 1/n , where 4ϽnϽ5͒, and the large scatter in absolute rate constants indicate diffusion-limited, rather than interface-limited mass transport, although both the facetnucleation step and the wetting-layer defects inhibit the cluster growth. The strain-induced energy barriers at the cluster bases prevent material addition, and thus growth of large clusters, facilitating the domination of growth mechanisms other than Ostwald ripening at temperatures below 700 K, leading to symmetric or positively skewed cluster-size distribution functions. The tendency towards negatively skewed and bimodal distributions at higher temperatures signaled the contribution of the ripening in accord with the Lifshitz-Slyozov-Wagner theory. Raising the temperature above 770 K leads to a gradual replacement of the huts by the ͗110͘-based macroscopic clusters. ͓S0163-1829͑97͒05039-X͔

Kinetically Self-Limiting Growth of Ge Islands on Si(001)

“…Here we examine the growth of Ge islands with ͕105͖ facets, the so-called "hut" clusters [7][8][9][10]. Since such Ge islands are the smallest observed on Si(001) [1,11], they are best suited as nanometer sized islands.…”

In situ transmission electron microscopy observations of the formation of self-assembled Ge islands on Si

“…To add complexity and interest to this story, under certain growth conditions a bimodal or trimodal distribution of island sizes has been observed: examples exist for the Ge/Si (Goldfarb et al, 1997a;Goryll et al, 1997;Kamins et al, 1997;Medeiros-Ribeiro et al, 1998) and InP/GaInP systems (Johansson et al, 1998;Lee et al, 1997;Ponchet et al, 1995;Reaves et al, 1996). In a beautiful STM study of the Ge/Si system (MedeirosRibeiro et al, 1998), it was shown that the two peaks in the island size distribution correspond to two different island shapes.…”

Is the Elongation of Ge Huts in the Low-Temperature Regime Governed by Kinetics?

“…1(d) using characteristic values for the Ge=Si001 system [22]. While the calculations indicate isotropic growth for s t e 0 100 nm ( 0 37 nm), real huts seem to begin elongation already at a stable nucleus size of about 5-8 nm [23], if at all [e.g., the island in Fig. 1(a) does not show any anisotropic elongation].…”