PLASMA ETCH MODELING USING OPTICAL-EMISSION SPECTROSCOPY

Plasma etching is often considered a yield limiter in the manufacturin g of Submicron integrated circuit devices. Much effort has been devote d to developing reliable models that relate the process outputs to var iations in real-time sensor signals. These models, called chamber stat e models, allow semiconductor manufacturers to predict etch behavior. In this article, we propose to use optical emission spectroscopy (OES) as a real-time sensor to quantify and predict the etching performance in an integrated circuit manufacturing line. This method is especiall y useful in plasma processing because it provides in situ and real-tim e analysis without disturbing the plasma or interfering with the proce ss. This study is based on an OES system that has been installed on an Applied Materials 5300 Centura dielectric etcher with a single optica l fiber connected from the reactor viewport to a spectrograph. A desig ned experiment was performed on oxide test wafers. Several etch charac teristics, including etch rate, within-wafer uniformity, and aspect-ra tio dependent etching (ARDE), were modeled in this study. Various mode ling techniques such as multivariate principle component analysis, and partial least squares were employed to relate the various OES signatu res to etching performance. The results show that 87% of etch rate var iation and more than 95% of the variation in within-wafer uniformity c an be explained by these models, although the OES signals can only exp lain 65% of the variation in ARDE. (C) 1996 American Vacuum Society.