Related Experiment Video
Updated: Jul 7, 2026
09:32
Cooling Rate Dependent Ellipsometry Measurements to Determine the Dynamics of Thin Glassy Films
Published on: January 26, 2016
Neural-net computing for interpretation of semiconductor film optical ellipsometry parameters
1Dept. of Electr. Eng. and Appl. Phys., Case Western Reserve Univ., Cleveland, OH.
IEEE Transactions on Neural Networks
|January 1, 1996
Summary
Neural networks efficiently invert optical ellipsometry data for semiconductor wafer analysis. This technique accurately determines film composition and thickness, crucial for molecular beam epitaxy processes.
Area of Science:
- Materials Science
- Optical Physics
- Computational Science
Background:
- Optical ellipsometry is valuable for monitoring semiconductor wafer fabrication processes.
- Calculating ellipsometry angles from film properties is straightforward.
- Inverting this relationship to determine film composition and thickness is mathematically challenging.
Purpose of the Study:
- To apply neural network computing for the inverse mapping of optical ellipsometry parameters.
- To develop an efficient method for interpreting ellipsometry data in terms of film properties.
Main Methods:
- Utilized a functional-link net, a type of neural network known for efficient function approximation.
- Applied the neural network to perform the inverse mapping of measured ellipsometry parameters.
Main Results:
- The functional-link net demonstrated high efficiency in approximating the complex inverse relationship.
- The neural network approach enabled a holistic interpretation of ellipsometry data.
Conclusions:
- Neural network computing, specifically functional-link nets, offers a powerful and efficient solution for inverting optical ellipsometry data.
- This method facilitates accurate determination of film composition and thickness in semiconductor wafer fabrication.
