Neural Network Modeling of Deposition Rate Characteristics of Low Temperature Silicon Nitride Deposited by Inner Two Parallel Coil Inductively Coupled Plasma Chemical Vapor Deposition

Abstract

In this study, a neural network model for silicon nitride (SiN) deposition process is proposed. SiN thin films were deposited by a direct inner two parallel inductively coupled plasma chemical vapor deposition (ICPCVD) system that can control the activated radical and charged species in plasma. This system can produce SiN thin films at low temperature for flexible displays. The input parameters considered for deposition conditions were the N2 gas flow rate, NH3 gas flow rate, and substrate temperature. These were varied within the ranges of 0-20 sccm, 0-20 sccm, and 100-300 degrees C, respectively. For those of input parameters and the output of deposition rate, we developed a back propagation neural network model with a pre-processor. It is shown that the model accuracy and learning speed of the proposed model are better than those of a conventional neural network model. In the experiments conducted, it was found that the deposition rate increased as the flow rates of ammonia (NH3) and nitrogen (N2) increased up to a certain amount. On the contrary, when the flow rates of NH3 and N2 went over a certain amount, the deposition rate decreased. It was also found that an increase in temperature resulted in an increase in the deposition rate.