A hierarchical model-based method for wafer level virtual metrology under process information deficiency

Abstract

Online inspection is one of the most critical processes of quality control in semiconductor manufacturing. The physical inspection methods for wafers are time-consuming and unable to achieve wafer level metrology. In order to improve production efficiency and expand inspection coverage, virtual metrology (VM) methods have recently received widespread attention; they utilize process parameters to estimate wafer metrology results. However, due to process drift and other reasons, the process information contained in real-time signal data (RTS data) used for VM modeling in industrial production is insufficient. This work proposed a hierarchical modeling method for machine learning-based virtual wafer metrology, leveraging RTS and post-process quality characteristics. The hierarchical model consists of an multiway principle analysis (MPCA) sub-model for RTS feature extracting and two separate long short-term memory (LSTM) networks for wafer-to-wafer dynamics in RTS and quality characteristics, respectively. A case study on the thickness VM of chemical vapor deposition thin film is conducted, and the proposed method has achieved better results than other methods in comparison.

Keywords:

• virtual metrology
• long short-term memory
• semiconductor manufacturing
• multiway principal analysis

Additional information

Funding

This work was supported by the financial support from National Natural Science Foundation of China under Grant NO.62173298.

Notes on contributors

Yu-Jun Liu

Yu-Jun Liu received her B.S. degree in Ocean College from Zhejiang University in 2015. She is currently a Ph.D. candidate in the College of Control Science and Engineering, Zheiang University. Her research focuses on virtual metrology in batch process.

Dong Ni

Dong Ni is a professor with the College of Control Science and Engineering, Zheiang University. He received a B.S. degree in Industria Automation from Zhejiang University in 2001 and a Ph.D. degree in Chemical Engineering from the University of California, Los Angeles in 2005. His research interests include artificial intelligence, multiscale system modeling and control, big data analytics, and their applications in semiconductor manufacturing and renewable energy.

Xiong Shao

Xiong Shao, Associate Division Director of Shanghai Huali Integrated Circuit Manufacturing Corporation, graduated from physics department of Shanghai University in 2002, specializing in data analysis and application support in IC manufacturing field.

Dan-Li Gong

Dan-Li Gong is the Director of Engineering Support Section of Huali Itegrated Circuit Manufacturing Co., Ltd. She graduated from the Mathematics Department of Shanghai University. Her main research interests are big data analysis, machine learning and artificial intelligence.

Jin-Jin Li

Jin-Jin Li received her master’s degree in 2020 from Tongji University. She is currently an engineer in Shanghai Huali Integrated Circuit Manufacturing Co. Ltd. Her research interests mainly focus on the Neural Networks.