Characterization of advanced shallow trench isolation (STI) CMP processes and consumables

The main objectives of this thesis work were to elucidate the correlation between key characteristics of Chemical Mechanical Planarization (CMP) slurry and their polishing performance in shallow trench isolation (STI) applications. More specifically, a range of physical and chemical properties of STI slurries such as particle type, particle size, slurry pH, polymer additives, zeta potential were characterized. Their influences on CMP performance were examined based on their removal rate, selectivity, and response to processing parameters. Furthermore, the dependence of these performance metrics on polishing platform were also investigated.; The slurry properties play a vital role in the polishing performance. The particle type influences the surface interaction and removal rate, particle size is significant in the removal rate of the film being planarized, the interaction between the abrasive particle and film surface is influenced by both the slurry pH and zeta potential and the polymer additives affect the oxide:nitride selectivity. These properties are typically studied in order to obtain a further understanding of their roles in the polishing performance. It is believed that the slurry pH, zeta potential and polymer additives are the most critical in the polishing performance for a more mechanical process. For the three slurries the abrasive particle is ceria and the particle size is similar. However, the slurry pH, zeta potential and polymer additives are not similar for each slurry. Each slurry behaves differently in terms of the polishing performance with these varying slurry properties.; The removal rate is influenced by the polishing pressure, platen/carrier speeds and polishing time. The oxide:nitride selectivity is important to the polishing of patterned wafers because if the selectivity is poor than the possibility of feature dishing and erosion is greater. It is important to study the removal rate as a function of varying process parameters in order to obtain the optimum polishing process conditions. Selectivity is also important in understanding the optimum polishing process conditions especially in terms of the over polishing window for a particular slurry. It is expected to obtain a decrease in removal rate as the polishing pressure is increased. During the study it was observed for STI-A the removal rate decreased as the polishing pressure increased.; Two different polishing platforms were used in this study. The rotary polishing platform consists of both the platen and carrier rotating in the same direction typically with the platen rotating at a slightly higher speed than that of the carrier. Polishing slurry is placed onto the platen and with the rotational movement of the platen and carrier the slurry is drawn between the pad and wafer surface. The orbital polishing platform is the other used in this study. Unlike the rotary platform the orbital platform consists of the platen moving in an orbital motion while the carrier in a rotational motion. The polishing slurry is passed through slurry ports in the pad directly to the wafer surface. It was believed that due to the different polishing platform motions and slurry transport mechanisms that these would influence the polishing performance. Therefore, we did not expect to achieve similar polishing results for both polishing platforms. It was observed that one of the slurries (STI-A) followed a non-Prestonian behavior for the rotational platform but a Prestonian behavior for the orbital platform.; Initial studies were performed using thermal oxide and nitride blanket wafers in order to obtain the respective within wafer non-uniformity and material removal rates for a variety of different polishing conditions. An IPEC/Westech 372M and a Mirra both rotational platform polishers and a Speedfam/IPEC 676 oribital platform polisher were used in this study. It was observed that one of the slurries showed a non-Prestonian behavior for the rotational platforms but no...