[19a-A36-1~7] 8.3 Plasma nanotechnology

9:30 AM - 9:45 AM JST(12:30 AM - 12:45 AM UTC)

[19a-A36-3] Machine Learning Prediction of SiO₂ Film Property from Optical Emission Spectroscopy in TEOS /O₂/Ar Plasma CVD

〇KUNIHIRO KAMATAKI¹, Sukma Fitriani², Yushi Sato¹, Yuma Yamamoto¹, Yousei Kurosaki¹, Daisuke Yamashita¹, Naoto Yamashita¹, Takamasa Okumura¹, Naho Itagaki¹, Kazunori Koga¹, Masaharu Shiratani¹ (1.Kyushu Univ. ISEE., 2.Kyushu Univ. IMI.)

Keywords:

machine learning,TEOS plasma CVD,SiO2 thin film

Finding the optimal process conditions in plasma processing incurs significant costs, making the utilization of machine learning a valuable approach. Plasma emission spectroscopy contains information such as the density of active species and electron temperature in the plasma, which can be difficult to handle. However, since it can be measured relatively easily and non-invasively, there is a need for methods that can effectively utilize this data. In this study, we attempted to predict the deposition rate in the formation of SiO2 thin films using machine learning based on the OES results of TEOS+O2+Ar plasma.