1 April 1998 Neural network approach to rapid thin film characterization
Author Affiliations +
A novel approach for thin film thickness and optical constant extraction from spectral reflectance data is presented here. This methodology combines the global minimization abilities of Adaptive Simulated Annealing with the high computational efficiency of Neural Networks to solve complex characterization problems in real time. The optical constants of many thin films such as Polysilicon are a function of the processing conditions and hence the real time measurement of these parameters could possibly be used in real time or run to run process control applications.
© (1998) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Nickhil H. Jakatdar, Nickhil H. Jakatdar, Xinhui Niu, Xinhui Niu, Costas J. Spanos, Costas J. Spanos, "Neural network approach to rapid thin film characterization", Proc. SPIE 3275, Flatness, Roughness, and Discrete Defects Characterization for Computer Disks, Wafers, and Flat Panel Displays II, (1 April 1998); doi: 10.1117/12.304402; https://doi.org/10.1117/12.304402

Back to Top