208. EFFECTS OF CHEMICAL MECHANICAL PLANARIZATION (CMP) SLURRY CHEMISTRY ON COPPER NANOHARDNESS AND ETCH RATES
Name: Robin V. Ihnfeldt
Grad Year: 2008
Chemical mechanical planarization (CMP) is used in integrated circuit manufacturing to remove excess material and provide a globally planarized wafer surface. The CMP process uses slurry containing nanometer-sized abrasive particles and chemical additives which produces a mechanical and chemical synergistic effect. Since copper is the interconnect of choice, the focus of our research is on copper CMP. The chemical additives in the slurry control the state of the copper (CuO, Cu+, etc.) on the surface of the wafer and in the slurry and also affect the dispersion characteristics of the abrasives. Previous work investigated the effects of chemical additives (glycine, H2O2, benzotriazole, etc.) on the colloidal behavior of an alumina dispersion and found that agglomeration typically increased with additives (1). In this study, the nanohardness and etch rates of copper films sputter-deposited onto a tantalum coating on silicon wafers were measured after exposure to the same slurry solutions. The nanohardness was measured using a nanomechanical test instrument (Hysitron, Inc.) with maximum applied loads between 50-3000 N. To obtain etch rates, the wafer pieces were weighed before and after immersion in the solution. It was found that for most cases, the hardness values were consistent with the formation of surface films as indicated by the equilibrium potential-pH diagrams. At low pH (<4) the hardness values are that of Cu metal or slightly higher. At higher pH values (>8) the hardness decreases to less than Cu metal as hydroxides form or to higher values than Cu metal as oxides form. Slurries with high etch rates (>10 nm/min) typically caused soft surface films to form, while slurries with low etch rates (<4 nm/min) caused surface films with hardness values near that of Cu metal or much higher when thick oxide layers formed. The nanohardness and etch rate measurements are incorporated into the model of CMP of Luo and Dornfeld (2) to predict material removal rate (MRR) and the predictions are compared to experimental copper CMP data.
References: 1. R. Ihnfeldt and J. B. Talbot, J. Electrochemical Soc., 153, G948 (2006). 2. J. Luo and D. Dornfeld, IEEE Trans. Semicond. Manuf., 14 2 (2001).