Seungchoun Choi

Graduate Student Researcher
Department of Mechanical Engineering
University of California Berkeley, CA, 94720

Mechanistic Modeling of Copper CMP to Explain Material Removal Behavior and Pattern Dependent Variability

Copper CMP has been a key enabling technology for the multilevel metallization of semiconductor interconnect. Despite its wide application, the material removal mechanism during the process has not yet been well understood and prediction of the result of the process has relied on empirical models. In addition, yield loss of the semiconductor manufacturing is increasingly related to the systemic error with the scaling of the devices and the interconnect, which can be addressed only by systemic understanding of the manufacturing processes. Previous CMP models have limited applicability because of their narrow range of analysis and empirical nature. Therefore, a mechanistic model for copper CMP that quantitatively predicts the material removal rates based on the fundamental understanding of the process was proposed in this research. The proposed model considered the synergism between chemical and mechanical aspects of the process. The model was also extended to explain pattern dependent variability during copper CMP. Experimental testing will be conducted to validate the proposed models.