Computational Study of SiC Halide Chemical Vapor Deposition System

Abstract

Halide CVD (HCVD) is recently employed to grow SiC epitaxial layers using SiCl4/C3H8/H2 mixtures in an effort to achieve high deposition rates. The introduction of the chlorinated species allows the formation of more stable species SiCl2 while maintaining high surface reactivity, thus avoiding the silicon gas phase nucleation that has been widely reported in conventional CVD process using SiH4/C3H8/H2. However, the difficulties in reducing defect density and controlling the electrical properties of the material present a significant technical obstacle for HCVD of SiC. In experimental growth, the electrical properties, defect densities and the growth rate of as-deposited SiC epitaxial films are, to a large extent, determined by processing parameters including temperature, pressure, flow rates of precursors and carrier gas. Optimization of growth conditions provides the opportunity to engineer films with desired film properties and qualities at high deposition rate but requires in-depth understanding the deposition process. In this study, we performed computational study to investigate the effects of main processing parameters in HCVD process on film growth. Numerical experiments were performed over a wide range of operational parameters to provide information on distributions of gas velocity, temperature, and chemical species’ concentrations in the reactor as well as the deposition rates on the substrate surface. Simulations were also carried out to address hot zone design and operational conditions.