• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2004년도 추계학술대회 논문집 Vol.17
• /
• pp.711-714
• /
• 2004

Chemical mechanical polishing (CMP) has been widely accepted for the global planarization of multi-layer structures in semiconductor manufacturing. However, it still has various problems to the CMP equipment, in particular, among the CMP components, process variables are very important parameters in determining the removal rate and non-uniformity. Using a design of experiment (DOE) approach, this study was performed investigating the interaction between the various parameters such as turntable and head speed, down force and back pressure during CMP. Using statistical analysis techniques, a better understanding of the interaction behavior between the various parameters and the effect on removal rate, no-uniformity and ETC (edge to center) is achieved.

• 한국전기전자재료학회논문지
• /
• 제18권1호
• /
• pp.24-29
• /
• 2005

Chemical mechanical polishing (CMP) has been widely accepted for the global planarization of multi-layer structures in semiconductor manufacturing. Copper has been the candidate metallization material for ultra-large scale integrated circuits (ULSIs), owing to its excellent electro-migration resistance and low electrical resistance. However, it still has various problems in copper CMP process. Thus, it is important to understand the effect of the process variables such as turntable speed, head speed, down force and back pressure are very important parameters that must be carefully formulated in order to achieve desired the removal rates and non-uniformity. Using a design of experiment (DOE) approach, this study was performed investigating the main effect of the variables and the interaction between the various parameters during CMP. A better understanding of the interaction behavior between the various parameters and the effect on removal rate, non-uniformity and ETC (edge to center) is achieved by using the statistical analysis techniques. In the experimental tests, the optimum parameters which were derived from the statistical analysis could be found for higher removal rate and lower non-uniformity through the above DOE results.