Numerical simulation and optimization of gallium nitride growth in MOCVD manufacturing process

Gallium nitride (GaN) is an attractive material for manufacturing light emitting diodes (LEDs) due to its wide band-gap and superb optoelectronic performance. The quality of GaN thin film determines the reliability and durability of LEDs. Metal-organic chemical vapor deposition (MOCVD) is a common technique used to fabricate high-quality GaN thin films. In this paper, GaN growth rate and uniformity in a vertical rotating disk MOCVD reactor are simulated based on a three-dimensional computational fluid dynamics (CFD) model via ANSYS-FLUENT. Simulation of transport phenomena and chemical kinetics in GaN growth process is performed with a reduced chemistry model which consists of 17 gas phase and 8 surface species are participating in 17 gas phase and 17 surface reactions. The influence of operating variables includes susceptor rotation rate, susceptor temperature, velocity inlet, the reactor pressure, and precursor concentrations V/III ratio on the GaN growth process is investigated. In the numerical simulation, factors that have a significant effect on the GaN growth rate and uniformity are identified. The response from simulation data with minimum error variance estimation is predicated using Kriging method. A surrogate model as a function of these parameters is generated to predict the factors for optimal growth rate and uniformity. In the final part, multi-objective optimization using a multi-objective genetic algorithm to generate the Pareto frontier of optimum growth rate and uniformity of GaN thin films is carried out. It has been shown that TMG flow rate and the reactor pressure have a significant effect on growth rate and uniformity of GaN thin films. The results reveal that the proposed optimization formulation can generate Pareto frontier of conflicting objectives, thus providing reliable solutions for decision makers.