Dynamic Time Warping for Time Series Data Analysis of STI Etching

Abstract

Numerical data of in-situ optical acquisition system were used in machine learning algorithm to evaluate an STI etching process in a dataset of more than 200 etched wafers processed during a year. Though overall recipe performance was according to specifications, the observed parameter fluctuations were characterized to check for correlations. Thus, K-neighbors regressor was used to extrapolate the values of etch depth uniformity for entire population of wafers based on limited amount of measurement points, while dynamic time wrapping was used to analyze time series datasets to get characteristics of minimum distance’s path between signals for all wafers. The metrics of distances cluster groups of wafers that appear to be in proximity. Out of eight analyzed channels, we observed the largest distance variability in the Si main etch for spectral reflectometer dataset, which we compared to optical emission spectroscopy channels. Based on correlation assessment of distances to the etch depth uniformity data, we showed that the selected approach is a good means to analyze time series datasets of e.g. dry etch processes for monitoring process stability.