Arc Detection in Plasma Etching Using Fuzzy Model and Dempster-Shafer Theory
| Autor: | Min-Woo Kim, Yi-Seul Han, Seung-Soo Han, Sung-Hwan Shin |
|---|---|
| Rok vydání: | 2012 |
| Předmět: | |
| Zdroj: | ECS Transactions. 44:1069-1074 |
| ISSN: | 1938-6737 1938-5862 |
| DOI: | 10.1149/1.3694430 |
| Popis: | In current semiconductor manufacturing process, plasma processes such as etch and CVD take the portion at least 40% throughout the integration processes. In etch and CVD using plasma, particles created during processes produce abnormal discharge with relatively large energy and light. This energy must be detected because this energy causes damage directly or indirectly to wafer. Using Optical Emission Spectroscopy (OES), it is possible to detect status changes of plasma and also to find arc generation. The OES is often used for analysis of plasma physics and chemistry in real-time. The OES system gathers a series of optical emission intensity of the plasma in the chamber. But OES has large amount data because of using the full range of wavelength. Therefore, it takes a long time to process the data. To solve this problem, a new method is proposed using Fuzzy model and Dempster-Shafer theory by utilizing equipment data from matching network. Six data from matching network such as Cap position, RF reflective powers, are used to make Fuzzy model for arc detection. Fuzzy model normalizes these matching network data between 0 and 1. Output of Fuzzy model is used as a reference to detect abnormal data from on-line data. Six matching network parameter are compared at the same time and Dempster-Shafer theory is used to make a decision whether current state is in arc state or not. The results show that the proposed method detects arc point accurately. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|