Offline Lithography Process Stability Control Method Based on Image

Abstract

In the lithography process, the stability of lithography conditions plays a crucial role in stabilizing the image quality on the wafer. The current approach for monitoring the stability of lithography conditions is to use a scatterometer and specially designed grating patterns to infer the actual lithography process conditions from the scattering profile curve. This method requires special scat-terometer tools and is sensitive to changes in the film stacking layer under the photoresist. To address the aforementioned challenges, this article proposes an image-based offline lithography process stability monitoring method. We propose the use of neural networks to extract feature information from CDSEM images under different lithography processes firstly. And match the