Microscopic view of epitaxial metal growth: nucleation and aggregation

TLDR: In this article, the authors reviewed the early stages of epitaxial growth and showed how the growth kinetics can be employed to create well-defined island morphologies and island arrays in a self-organization process.

About: This article is published in Surface Science Reports.The article was published on 1998-01-01 and is currently open access. It has received 851 citations till now. The article focuses on the topics: Nucleation & Island growth.

Figures

Fig. 36. Preferred attachment towards A-steps is intrinsic for all surfaces with hexagonal symmetry. (a) Model showing how atoms from nearest neighbor sites are funneled preferentially towards A-steps. This asymmetry is caused by hcp-sites located in phase with the step atoms for A-steps, whereas they are out-of-phase and also far apart at B-steps. (b) EMT calculations for diffusion around a heptamer as in Fig. 34(a), for a number of systems. They all show the asymmetry around the hcp site located before an A-step whereas corner diffusion is system speci®c. The energy barriers were scaled by the EMT-values for terrace diffusion of the respective system.

Fig. 19. Anisotropic corner rounding as the dominant mechanism giving rise to elongated islands on fcc(1 1 0). (a) Ball model showing the processes considered. Their barriers are tabulated. (b) Simulated aspect ratios (®lled symbols) from three parameter sets I, II, and III in comparison with experimental results (open symbols) (from Ref. [216]).

Fig. 13. Deviations from scaling and post-deposition mobility studied through comparison of calculated to experimental island densities. (a) The transition from dynamic nucleation (with i 1) to post-growth with decreasing temperature is well reproduced for the system Cu=Ni(1 0 0) by integrating rate equations (lattice approximation) until 2 h after deposition (from Ref. [77]). (b) Distinction between deviation from scaling and the effect of post-deposition mobility in comparison to experimental results for Ag=Pt. (c) Effect of easy attachment to islands equally explains the experiment for Ag=Pt. (b) and (c) KMC performed for dendritic islands with the diffusion parameters for Ag=Pt(1 1 1). (d) EMT calculation for Ag=Pt(1 1 1) showing the potential energy for an atom approaching a one- and two-fold site at an island (from Ref. [85]).

Fig. 6. Variation of the saturation island density with temperature for deposition of 0.12 ML Ag onto a Pt(1 1 1) surface at 80, 95, and 110 K, respectively (F 1:1 10ÿ3 ML=s).

Fig. 30. Classical DLA aggregate of 3600 particles on a square lattice (from Ref. [15]).

Fig. 8. STM images showing Ostwald ripening in two dimensions as a way to create compact 2D Ag islands with their size being well de®ned by the annealing temperature. The starting population of mostly dimers and trimers was produced via deposition of 0.1 ML Ag onto Pt(1 1 1) at 50 K. The onset of Ostwald ripening is marked by a sharp transition of constant to exponentially increasing island sizes as a function of annealing temperature. From the temperature and the annealing period the dimer bond energy and the temperature regime for i 1 can both be inferred (from Refs. [22,59]).