Presentation Information

[Tu-P2-G-04]Optical Metrology of Semiconductor Nanowires Beyond the Diffraction Limit using High-Throughput Correlated Microscopy

〇Greg Chu1, Zeinab Tirandaz1, Ralf Mouthaan2, Hoe Tan3, Chennupati Jagadish3, Hannah Joyce1, Jack Alexander-Webber1 (1. University of Cambridge (UK), 2. University of Adelaide (Australia), 3. The Australian National University (Australia))
By correlating optical images with SEM images, we are able to train a machine learning model that can predict a nanowire diameter from optical images alone within 5% (<13nm) mean-relative error, reducing cost, increasing throughput, without compromising accuracy.