Optimizing Temperature and Flow Fields of 4H-SiC Epitaxial Growth by Integrating CFD Simulation with Multi-objective Particle Swarm Optimization

The silicon carbide (SiC) epitaxial growth process is crucial in chip manufacturing. The distribution of the flow and temperature fields in the reactor chamber influences the epitaxial layer uniformity. Therefore, this study optimizes the distribution of the flow and temperature fields inside the reactor to enhance the quality of the epitaxial layer. COMSOL Multiphysics is used to model the horizontal chemical vapor deposition (CVD) reactor chamber, and the flow and temperature fields inside the reactor chamber are analyzed. Factors influencing the uniformity of flow field distribution include the reactant gas distribution and the gas-inlet tunnel’s diameter and position. The flow field uniformity is represented by the relative standard deviation of the velocity. Parameters impacting the temperature field uniformity include the position and pitch of the heating coil and the graphite column width. The heating efficiency of the substrate and temperature uniformity are expressed by the average temperature and standard deviation of the temperature, respectively. Support vector machine (SVM) is used to establish the relationship between design variables and the objective function, and the multi-objective particle swarm optimization (MOPSO) algorithm is used to optimize the reactor. The proposed approach improves the uniformity of the flow and temperature fields and the heating efficiency of the substrate.