Optimization of Substrate Temperature for Uniform Graphene Synthesis by Numerical Simulation and Machine Learning

High uniformity graphene has extensive application prospect in many important fields due to its excellent features. During large‐area graphene synthesis by chemical vapor deposition, the optimization of the substrate temperature can improve the uniformity of graphene. Here, machine learning is used to design and optimize the substrate surface temperature for uniform graphene deposition. The computational fluid dynamics simulations based on a developed computational model are first performed to obtain the training data for machine learning, such as the gas temperature, velocity, concentrations, etc. Then, the neural network model is used to optimize the substrate temperature using the simulated data. It is found that the high accuracy is achieved through the validation of testing set. The optimal substrate temperature distribution is finally obtained, in which the carbon deposition rate and its uniformity are optimized to the specified values. This work focuses on substrate temperature optimization for uniform graphene synthesis by combination of computational fluid dynamics (CFD) simulation and machine learning. The main processes include: 1) initial parameters’ selection to generate the temperature distributions for CFD simulations; 2) construction of a data set for neural network training; 3) prediction by NNs model; 4) output of the optimal parameters for uniform graphene synthesis.