Corey Clark

Graduation Semester and Year




Document Type


Degree Name

Doctor of Philosophy in Electrical Engineering


Electrical Engineering

First Advisor

Choong-Un Kim


Modeling of deposition processes has become of extreme importance due to the small scale of devices and features as well as the reduction of time to market required by industry. Current modeling procedures have focused on individual aspects of growth as well as made assumptions that cause simplification of the problem to the point the models usefulness is limited. The Kinetic Monte Carlo (KMC) model developed in this work combines rate transitions that have been commonly found in KMC simulation along with energy density equations that help explain the transitions and formations of island in alloy deposition. Specifically the model developed in this study, analyzes both the surface energy and strain energy of the film, which are incorporated to show the dependence of strain relaxation to surface energy during island formation. The developed model also incorporates the anisotropy of crystalline structures to accommodate for the changes in growth rate and morphology based upon crystal orientation. This leads to a more versatile model that will accommodate multiple material sets as well allow for quick simulation results for the development of new devices. Work was also done in increasing the randomness of site selection while minimizing errors due to standard uniform number generators. The developed KMC model incorporates a pseudo random number generator for the purpose of site selection, which reduce the amount of cluster processing that can occur with random number generators. A focus was also placed on the ability to describe flux distributions that are not commonly found in semiconductor device manufacturing. This was done to allow for the expansion of this model into non-planar environments that might be found in industries such as MEMS/NEMS. This extension also allows for evaluation of non-planar deposition process such as via deposition. The expansion done in this model allow for a wider variety of applications with in the semiconductor field. Accounting for crystal orientation allows for increased accuracy as well as further insight into its affect on island formation within alloy growth. The changes in random surface site selection during simulation have helped reduce cluster processing which allows for an accurate depiction of surface morphology.


Electrical and Computer Engineering | Engineering


Degree granted by The University of Texas at Arlington