Modeling and simulation of coverage and film properties in deposition process on large-scale pattern using statistical ensemble method

Abstract

This study modeled deposition processes using statistical ensemble and feature-scale voxel methods to predict the coverage and film properties on a large-scale pattern for the first time. Certain new concepts, such as the use of probabilities to express physical and chemical phenomena occurring on the surface, interaction among voxels, and super particles for fast calculations, were introduced. Further, the model was used to analyze the experimental characteristic variation between the morphology and film density of SiN in low-temperature plasma-enhanced chemical vapor deposition using a SiH4/NH3/N2 gas mixture with different SiH4 flow rates, which has not been found in SiO2 films. The simulation results demonstrated the effect of the precursor mass in the gas phase on surface migration and morphology formation. In addition, a short residence time was required to prevent the generation of large and heavy precursors to realize good SiN coverage and film properties at a low temperature.