Stimulated by our investigations of minimum feature size in
laser-focused atomic deposition we have undertaken a series of calculations to explore the role of surface diffusion
in determining line width [EPG Pub# 703]. Starting with a flux
distribution calculated from atom optical considerations, we have applied
three different growth models to help provide insight into the observed phenomenon that deposited Cr lines tend to
broaden as more material is added.
Our first model, in which atoms were allowed to move one or two sites after landing, based on a thermally
driven rate coefficient, gave line profiles that did not match our observations at all-lines did not
grow in width so much as develop flat tops, and the shape was extremely sensitive to deposition temperature
(something not seen in experiments).
The second two models invoked ballistic deposition, in which atoms are allowed to stick to whatever local
surface (horizontal or vertical) they encounter first. This leads to fractal-like growth, and a marked
increase in line width as more material is added. While ballistic deposition alone led to an unrealistic
number of voids in the deposited material, we found that if we added a small amount of relaxation (that is,
a probability of moving to a neighboring site with more neighbors), we could get behavior that showed a
good qualitative match to what was observed in the experiments.
While these models give some understanding of the role played by surface growth in laser-focused atomic
deposition, much work still needs to be done to explore this subject fully. New depositions need to be
carried out in ultrahigh vacuum to eliminate the possible influence of contaminants, and more sophisticated
models need to be considered. With these improvements, laser-focused atomic deposition has the potential
to become a useful tool for studying surface growth.

Calculated deposition profiles based on (a) ballistic deposition alone and (b) ballistic deposition with relaxation.
Online: August 2000
Last update: February 2008
Website Comments:egpwebmaster@nist.gov