Ellipsometric characterization of photo-resist gratings using artificial neural network

摘要

In this work, we report on the control of patterned photoresist structures used as a mask in the photolithographic process. This control is crucial in order to detect any defect before the final etching step on the substrate or sublayer. Scatterometry is an optical method that has true potential for critical dimension and profile analysis.rnSpectroscopic ellipsometry is employed to measure the diffracted signature, and followed by a neural treatement to extract the geometrical parameters. Experimental results are validated on 2-μm-period gratings, with various geometrical profiles, transfered on a glass substrate.