Computational Lithography for Silicon Photonics Design
Year: 2019
Type: article
Abstract: We demonstrate a method to predict the photolithography effects of a 193 nm deep ultraviolet (DUV) process, and we build a lithography model using physical measurements on a fabricated test pattern. Our model accurately predicts the proximity and smoothing effects that are characteristic of DUV lithography. We verify the accuracy of the model by visually inspecting the fabricated test patterns and comparing them to the predicted ones. Additionally, using predicted shapes, we compare experimental measurements on a benchmark device (the contra-directional coupler) against those of our predicted, computed responses. Our predictions show good agreement with the fabricated results using both verification methods. Lastly, we illustrate the design-for-manufacturability enabled by our model by demonstrating a design correction method for the contra-directional coupler that improves its compatibility with the DUV process.
Institution: University of British Columbia
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
Award: G734200333
OpenAlex ID: W3000397606
