Share Email Print

Proceedings Paper

Large-area optical proximity correction using pattern-based corrections
Author(s): David M. Newmark; Sheila Vaidya; Jakub Segen; Andrew R. Neureuther
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Resolution enhancement techniques have been explored extensively in the last few years to reliably extend optical lithography to smaller features. In fact, remarkable depth of focus and resolution enhancements have been achieved for certain types of features. However, proximity effects can render these enhancements irrelevant because they can cause such severe linewidth changes that even in-focus lines are incorrectly sized. Other researchers have attempted to solve this problem using a wide variety of different approaches. Their methods have the common disadvantage that they try to correct all aspects of every pattern on the mask when it may be necessary to optimize only a small subset of patterns. Our technique, pattern recognition with polynomial corrections is developed for correcting only certain patterns. The development of this system led to several important results. First, a simple local bias system, which was developed to independently bias small patterns, shows that bias solutions quickly converge in a few iterations. Also, increasing the complexity by more finely dividing the edge segments does not increase the bias time for the patterns. The second important insight is related to the pattern recognition system. As in the other techniques, a proximity window is moved through the layout. However, in the pattern matching approach, a pattern matching zone within the window is extracted and used to find the corresponding pattern in the pattern library. The bias of the peripheral features outside the pattern matching zone, but within the proximity window are incorporated into the bias for the pattern by way of an inter-feature proximity correction polynomial. An important insight discovered with regard to the correction function is that the peripheral features can be subdivided into pixels whose bias contributions can be linearly superimposed to give an accurate approximation to the bias of the main feature.

Paper Details

Date Published: 7 December 1994
PDF: 13 pages
Proc. SPIE 2322, 14th Annual BACUS Symposium on Photomask Technology and Management, (7 December 1994); doi: 10.1117/12.195835
Show Author Affiliations
David M. Newmark, Univ. of California/Berkeley (United States)
Sheila Vaidya, AT&T Bell Labs. (United States)
Jakub Segen, AT&T Bell Labs. (United States)
Andrew R. Neureuther, Univ. of California/Berkeley (United States)

Published in SPIE Proceedings Vol. 2322:
14th Annual BACUS Symposium on Photomask Technology and Management
William L. Brodsky; Gilbert V. Shelden, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?