1 October 1997 Wavelet algorithms for in-situ monitoring of thin-film deposition
Author Affiliations +
Abstract
Fabrication of high performance optical filters are typically long deposition runs. Successful deposition requires the monitoring and correction for long term process variability due to source depletion and equipment degradation as well as control of immediate parameters such as deposition rates and substrate temperature. Multiple resolution analysis of insitu sensor data using wavelet transforms provides the means of extracting both short term and long term trends from a common data stream. This paper presents a method for monitoring the fabrication of discrete and Rugate filters using the Haar transform. This technique begins by describing the average of the time series, and then successively resolves the time series into more detailed layers. The lowest resolution coefficients describe optical density growth and absorption. The more detailed layers describe systematic and periodic variations due to the growth of optical thickness. Separate analysis of data for different time regimes provides insight into different physical parameters. Least square fit to the low order terms relate to parameters such as film absorption and growth of optical density. Windowed Fourier transform of periodic structure in the coefficients of the detailed layer provides insight into a number of performance parameters such as the period and amplitude of the optical thickness cycle which is characteristic of Rugate design.
© (1997) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Thomas D. Rahmlow, Jeanne E. Lazo-Wasem, David A. Rahmlow, "Wavelet algorithms for in-situ monitoring of thin-film deposition", Proc. SPIE 3133, Optical Thin Films V: New Developments, (1 October 1997); doi: 10.1117/12.279105; https://doi.org/10.1117/12.279105
PROCEEDINGS
12 PAGES


SHARE
Back to Top