Simulations of thin film growth on patterned substrates

Patterned substrate growth is a technique that layers of thin film are grown on a substrate with a predetermined pattern. A simple model is used to study patterned growth process with two different types of pattern: flat pattern and periodic pattern. The goals are to determine growth conditions that...

Full description

Saved in:
Bibliographic Details
Main Authors: Patcha Chatraphorn, Sojiphong Chatraphorn
Other Authors: Chulalongkorn University. Faculty of Science
Format: Technical Report
Language:English
Published: Chulalongkorn University 2008
Subjects:
Online Access:http://cuir.car.chula.ac.th/handle/123456789/6722
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Chulalongkorn University
Language: English
Description
Summary:Patterned substrate growth is a technique that layers of thin film are grown on a substrate with a predetermined pattern. A simple model is used to study patterned growth process with two different types of pattern: flat pattern and periodic pattern. The goals are to determine growth conditions that enable the grown film to reproduce the original pattern, and to determine how much of the original patterns survive up to a specific time. The persistence probability is used to determine fractions of survived pattern. We found that in flat patterned growth, a high substrate temperature which results in a long surface diffusion length of moving atoms can help increase the persistence probability of the pattern. If the substrate temperature is high enough, the film is grown in layer-by-layer mode and the flat pattern persists for a long time. In periodic pattern growth, long surface diffusion length helps with the smoothness of the flat parts of the pattern but destroys the outline shape of the pattern. We found that the substrate temperature has to be a moderate value, not too low and not too high. The optimal value depends on the size of the pattern. A pattern with a bigger feature size can persist for a longer period of time. Finally, we suggest a modified definition of the persistence probability in order to have a probability that agrees better with the simulated morphology.