Semi-empirical material removal rate distribution model for SiO2 chemical mechanical polishing (CMP) processes

A novel semi-empirical model was developed for predicting the material removal rate (MRR) during chemical mechanical polishing (CMP) based on the following assumptions: plastic contact at the wafer–particle interface, elastic contact at the pad–particle interface, a particle size distribution, and a randomly distributed surface roughness of the polishing pad. The proposed model incorporates the effects of particle size, concentration, and distribution, as well as the slurry flow rate, pad surface topography, material properties, and chemical reactions during the silicon dioxide (SiO2) CMP. To obtain the unknown parameters and ensure the validity of the model, a SiO2 CMP experiment was conducted by using various-sized CMP slurries. The spatial distribution of the MRRs is expressed with respect to the normal contact stress distribution and the relative velocity distribution. The proposed MRR model can be used for the development of a CMP simulator, the optimization of CMP process parameters, and the design of next-generation CMP machines.