Deposition, diffusion and aggregation on percolation substrates were investigated via computer simulation. The degree of non-uniformity of the substrate was described by the occupied probability, p, with which the percolation was generated; where p took values ranging from a threshold value, up to unity. Blocked sites in percolation represented defects in the substrates. Interactions between defects and deposited particles were incorporated by introducing a sticking coefficient, s. Inert defects (s = 0) hindered deposited particles from diffusing on the substrates. As p decreased from unity to the threshold value, the morphology of the aggregates varied from deposition, diffusion and aggregation patterns on uniform substrates to few- and zig-zag branch patterns. Active defects (s ≠ 0) played a role in absorbing deposited particles. Upon reducing p from 1 to the htreshold value, the pattern of aggregates changed from deposition, diffusion and aggregation on uniform substrates to a site-percolation like pattern (s = 1) or a dispersed small-island one (0 < s < 1) on critical percolation substrates. A rapid increase in the fractal dimension, D, of aggregates appeared in the D versus p curve; corresponding to a transition of morphology from a pattern dominated by defects, to one controlled by diffusion. The simulations showed that the Honda-Toyoki-Matsushita relationship was reasonable for growth which was controlled by defect-hindering diffusion in fractional spaces.

Deposition, Diffusion and Aggregation on Leath Percolations - a Model for Nanostructure Growth on Non-Uniform Substrates. Z.J.Tan, X.W.Zou, S.Y.Huang, Z.Z.Jin: Physical Review B, 2002, 65[23], 235403 (5pp)