Abstract

Thin film characterization is a necessary step in the semiconductor industry and nanodevice fabrication. In this work, we report a learning-assisted method to conduct the measurement based on a multi-angle polarized microscopy. By illuminating the film with a tightly focused vectorial beam with space-polarization nonseparability, the angle-dependent reflection coefficients are encoded into the reflected intensity distribution. The measurement is then transformed into an optimization problem aiming at minimizing the discrepancy between measured and simulated image features. The proposed approach is validated by numerical simulation and experimental measurements. As the method can be easily implemented with a conventional microscope, it provides a low cost solution to measure film parameters with a high spatial resolution and time efficiency.

© 2024 Optica Publishing Group

Rapid ellipsometric imaging characterization of nanocomposite films with an artificial neural network

Patrick Kfoury, Yann Battie, Aotmane En Naciri, Michel Voue, and Nouari Chaoui
Opt. Lett. 49(3) 574-577 (2024)

Poling-assisted hydrofluoric acid wet etching of thin-film lithium niobate

Simin Yu, Hui Kang, Xiaoqin Shen, Yibo Xue, Wenjie Wan, Changling Zou, Baile Chen, and Juanjuan Lu
Opt. Lett. 49(4) 854-857 (2024)

Deep-learning-assisted snapshot optical tomography for microscopic volume prediction: a simulation study

Andrew Richard Abramczyk and Yongjin Sung
Opt. Lett. 49(2) 302-305 (2024)

High performance polarizer on thin-film lithium niobate with width-tapered Euler bending

Weixi Liu, Chengfeng Wen, Chenkun Lei, Lijia Song, Daoxin Dai, and Yaocheng Shi
Opt. Lett. 49(9) 2337-2340 (2024)

Deep learning empowers photothermal microscopy with super-resolution capabilities

Yonghui Wang, Zhuoyan Yue, Fei Wang, Peng Song, and Junyan Liu
Opt. Lett. 49(8) 1957-1960 (2024)