Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


Breaking the 0.1 Barrier in Optical Microlithography via Signal Processing

The current predicted resolution limit for optical lithography is 0.13 m, using 193 nm laser sources and resists that are still under development. Beyond that, to quote Gordon Moore, there seem to be only three equally unappealing options: X-rays, e-beam, and deep UV (with mirrors instead of lenses). It appears that signal processing ideas can allow us to go beyond the so-called "0.1 barrier". The first steps in this direction were taken by Marc Levenson and others at IBM who showed that one could get well-separated images for two lines spaced closer than the conventional Rayleigh limits by adding a 180 phase shift to one of the lines. This idea, extended to more complex patterns, is being explored by many semiconductor manufacturing organizations as a way of extending the resolution of existing optical exposure systems. However, for all but very simple patterns, mask design is currently done in a laborious and empirical way, with in fact no guarantee that a suitable phase-assignment even exists. In this talk, we shall describe a systematic procedure based on the (approximate) solution of a nonlinear inverse problem - going from a desired pattern on a wafer to the specification of a mask that will yield this result by appropriately compensating for the effects of the optical projection system. Some experimental results will be displayed showing satisfactory printing of lines at a spacing 3 times as close as predicted by the nominal resolution, e.g., 0.11-0.13 m spacings using a system with 0.35 m nominal resolution. Comparable reductions have been obtained with systems of higher nominal resolution, providing hope that refractive-optical lithography could be used well into the next decade. This talk is based on work done with Dr. Y.C. Pati and Dr. Yao-Ting Wang at Stanford University and Numerical Technologies, Inc.

Type of Seminar:
Public Seminar
Thomas Kailath
Department of Electrical Engineering, Stanford University, Stanford, CA 94305-9510
Jul 07, 1999   15:15

ETH-Zentrum, Gloriastrasse 35, ETZ E6, 8006 Zürich
Contact Person:

Manfred Morari
No downloadable files available.
Biographical Sketch: