High-fidelity curvilinear SRAF method adopting distance-versus-angle-signature

  • Weichen Huang
  • , Yanqiu Li*
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Curvilinear sub-resolution assist features (SRAF) can significantly improve the fidelity of print image in curvilinear optical proximity correction (OPC). However, conventional inverse lithography technology (ILT)-based methods for curvilinear SRAF generation consume substantial computational resources. This paper proposes an efficient distance-versus-angle-signature (DVAS) based method for curvilinear SRAF placement. The proposed method leverages second-order derivative analysis of DVAS to rapidly identify SRAF insertion points. Furthermore, this method employs B-spline to connect these points, and applies morphological dilation operation to ensure constant-width SRAF generation. Simulation results demonstrate that the proposed method effectively places the SRAF and improves imaging performance.

Original languageEnglish
Title of host publicationNinth International Workshop on Advanced Patterning Solutions, IWAPS 2025
EditorsYayi Wei, Tianchun Ye
PublisherSPIE
ISBN (Electronic)9781510699298
DOIs
Publication statusPublished - 3 Dec 2025
Externally publishedYes
Event9th International Workshop on Advanced Patterning Solutions, IWAPS 2025 - Shenzhen, China
Duration: 14 Oct 202515 Oct 2025

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume13991
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference9th International Workshop on Advanced Patterning Solutions, IWAPS 2025
Country/TerritoryChina
CityShenzhen
Period14/10/2515/10/25

Keywords

  • Curvilinear mask
  • curvilinear optical proximity correction
  • curvilinear sub-resolution assist features

Fingerprint

Dive into the research topics of 'High-fidelity curvilinear SRAF method adopting distance-versus-angle-signature'. Together they form a unique fingerprint.

Cite this