Plasma information-based virtual metrology (PI-VM) and mass production process control

Abstract

© 2022, The Korean Physical Society.In this paper, we review the development of plasma engineering technology that improves dramatically the production efficiency of OLED (organic light-emitting diode) displays and semiconductor manufacturing by utilizing a process monitoring methodology based on the physical domain knowledge. The domain knowledge consists of plasma-heating and sheath physics, plasma chemistry and plasma-material surface reaction kinetics, and plasma diagnostics. Based on this, a plasma information-based virtual metrology (PI-VM) algorithm was developed drastically enhanced process prediction performance by parameterizing plasma information (PI) which can trace the states of processing plasmas. PI-VM has superior process prediction accuracy compared to the classical statistics-based virtual metrologies. The developed PI-VM algorithms adopted for practical processing issues such as the control and management of the OLED-display mass production demonstrated savings of approximately 25% of the yield loss over the past 5 years. This improvement was achieved with the development of FDC (fault detection and classification) and APC (advanced process control) logic, which can be developed through the analysis of the physical characteristics of the feature parameters used in PI-VM with the evaluation of their contributions and their correlations to the processing results. PI-VM provides leverage that can be applied in the development of process equipment and factory automation technologies.