Shoji Seta

Profile Control of SiO2 Trench Etching for Damascene Interconnection Process

The purpose of the present study is to reveal a microtrench generation mechanism in the damascene trench etch process for SiO2 film. Experiments are discussed for the CO gas flow and the pressure change, employing a C4F8/CO magnetron etch system. Increasing the CO gas flow or the pressure prevents the microtrench generation. Moreover, the microtrench ratio A/B that is defined by the depth at the edge of the trench bottom, A and the trench depth at the center of the trench bottom, B becomes higher in accordance with increasing larger space size. In previous reports, the main effect of CO gas and pressure increase is identified as decreasing CFx radical density by dilution of C4F8 or by increasing the residence time. The reduction of CFx radical flux may prevent microtrench generation by decreasing the formation of fluorocarbon film at the trench bottom. These results suggest that the microtrench generation is caused by thicker formation of fluorocarbon films at the center of the trench bottom which has a larger solid angle than at the edge of the trench bottom.

Microtrench Generation in SiO2 Trench Etching for Damascene Interconnection Process

The mechanism of microtrench generation in SiO2 trench etching in fluorocarbon gas chemistry is presented using the magnetron etch system under a 40–80 mTorr pressure condition. In the previous study, we pointed out that the microtrench is caused by the etch rate increase at the trench bottom edge where the fluorocarbon film is thin, because the thicker fluorocarbon film was formed more easily at the center of the trench bottom where the solid angle is larger than at the edge of the trench bottom. In this study, the experimental condition is extended to lower pressures. The microtrench ratio becomes higher in accordance with a larger pattern size at pressures higher than 40 mTorr. However, the pattern dependence of the microtrench ratio is reversed at 20 mTorr. This is explained by the effect of a constant residence time, the dilute gas effect, and the inverse-reactive ion etching (RIE) lag which is observed at 20 mTorr. In conclusion, the microtrench formed by the shadowing effect for fluorocarbon radicals is offset by increasing the bombarded ions at the trench bottom with the decrease of pressure to a condition such as 20 mTorr pressure.

Dual Damascene Etching Process Using Sacrificial Spin-on-Glass Film

The dual damascene (DD) formation process for sub 0.2 µm feature size devices has been investigated. The via-first DD process, where via holes are first formed, followed by trench formation, used the antireflection film in via holes after the trench pattern lithography process. In subsequential trench etch process, crownlike etch residues of SiO2 were formed around via holes due to etch inhibiting effect of antireflection films in via holes. A spin-on-glass (SOG) film is filled in via holes as a sacrificial film before trench pattern lithography was used. It could eliminate the crownlike etch residues because of the absence of an antireflection film in via holes. Moreover, it could give via holes a rounded profile at their top edges. This DD process used for fabricating films with no etch residues and rounded via holes reduces the aspect ratio of via holes and was confirmed to be effective for subsequential metal filling process.

Mechanism of Microtrench Generation in Etching of Wiring Trench on SiO2 Layer: Proposal of Simulation Model using High-Pressure Etching Gas

In this study, we propose a simulation model to clarify the mechanism of microtrench generation in the etching process of a wiring trench on a silicon dioxide (SiO2) layer using a high-pressure etching gas. Furthermore, the validity of the simulation model is clarified by a comparison with the experimental result. The trench etching process can be expressed by a transportation model outside the trench where radicals and ions are transported from the plasma to the surface of the SiO2 layer, a transportation model inside the trench where radicals and ions are transported to the trench bottom, and a surface reaction model for the SiO2 layer where the etching behavior of fluorine reaching the trench bottom is modeled. In the transportation model outside the trench, the measurement values for the density of reactive radical species in the plasma, electron temperature, and electron density should be used. In the transportation model inside the trench, the insulation layer characteristic and the behavior of radicals and ions in the trench with the aspect ratio of the trench should be considered. The surface reaction model of the SiO2 layer can use the etching rate for the blanket of the SiO2 layer and the real ion energy should be considered. Moreover, the simulation results obtained using this model correspond well to the experimental results, and the validity of the proposed model is shown.

Mechanism of Microtrench Generation in Etching of Wiring Trench on SiO2 Layer

In this study, we reveal the effects of various factors that influence microtrench generation using a high-precision simulation model of etching of wiring trench on a SiO2 layer. It is considered that the behaviors of ions and fluorine that contribute to etching, as well as the CO/C4F8 + CO flow rate ratio, have an effect on the microtrench generation. The number of ions on the trench bottom that contributes to the etching rate is larger at the side edges than at the center. This indicates that the trench bottom becomes a microtrench profile. Otherwise, the number of fluorine absorbed and used for etching is larger at the center of the trench bottom than at the side edges, regardless of the incident angle of radicals. This indicates that the trench bottom becomes a round profile. Moreover, an increase in CO/C4F8 + CO flow rate ratio decreases the number of fluorine at the side edges compared with that at the center. This indicates that the trench bottom becomes a more round profile. These simulation results reveal the mechanism of microtrench generation in the wiring trench of a SiO2 layer. Furthermore, we clarify guidelines for setting the conditions of the microtrench-free etching of a 200 nm trench bottom on a SiO2 layer.