Data-Driven Analysis of High-Temperature Fluorocarbon Plasma for Semiconductor Processing

Abstract

The semiconductor industry increasingly relies on high aspect ratio etching facilitated by Amorphous Carbon Layer (ACL) masks for advanced 3D-NAND and DRAM technologies. However, carbon contamination in ACL deposition chambers necessitates effective fluorine-based plasma cleaning. This study employs a high-temperature inductively coupled plasma (ICP) system and Time-of-Flight Mass Spectrometry (ToF-MS) to analyze gas species variations under different process conditions. We applied Principal Component Analysis (PCA) and Non-negative Matrix Factorization (NMF) to identify key gas species, and used the First-Order Plus Dead Time (FOPDT) model to quantify dynamic changes in gas signals. Our analysis revealed the formation of COF3 at high gas temperatures and plasma power levels, indicating the presence of additional reaction pathways under these conditions. This study provides a comprehensive understanding of high-temperature plasma interactions and suggests new strategies for optimizing ACL processes in semiconductor manufacturing.