Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

E-Beam Proximity Effect Dose Correction Using Computer Proportional to the Number of Shapes to be Corrected

IP.com Disclosure Number: IPCOM000049091D
Original Publication Date: 1982-May-01
Included in the Prior Art Database: 2005-Feb-09
Document File: 3 page(s) / 40K

Publishing Venue

IBM

Related People

Grobman, WD: AUTHOR

Abstract

The ideal incident electron dose distribution is calculated which will yield the desired ideal exposure distribution for each individual shape in a pattern (not taking into account adjacent shapes). Then these ideal incident electron dose distributions for each shape are edded together to yield an ideal total incident electron dose distribution for the entire pattern. The appropriate dose for each shape is obtained by averaging the total dose distribution within each shape. Large changes in dose distribution occurring within a shape indicate that the shape should be partitioned.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

E-Beam Proximity Effect Dose Correction Using Computer Proportional to the Number of Shapes to be Corrected

The ideal incident electron dose distribution is calculated which will yield the desired ideal exposure distribution for each individual shape in a pattern (not taking into account adjacent shapes). Then these ideal incident electron dose distributions for each shape are edded together to yield an ideal total incident electron dose distribution for the entire pattern. The appropriate dose for each shape is obtained by averaging the total dose distribution within each shape. Large changes in dose distribution occurring within a shape indicate that the shape should be partitioned.

Given a set of shapes with incident dose I(i) (r) within each shape, the resultant dose is R (r), where r is a vector in the plane and Kern (1) has solved this problem for a single shape, as illustrated in Fig. 1, finding that I(i) (x) was required to obtain a desired R(i) (x) for a given isolated shape i. (In this discussion, we illustrate all ideas in terms of long parallel lines in the y-direction, so that the problem becomes one dimensional, in x, and can be demonstrated more clearly.) This article (1), however did not describe a self-consistent correction for a set of shapes. Thus, Kern solved(see original) for the case in which R(i) (x) is a piecewise constant function. The result for I(i) (x) in Fig. 1 is a function which is negative outside the bounds of the original shape. Kern then proposed making these negative going parts of the function I(i) equal to zero to get an approximate "best" incident electron dose solution.

In order to get a self consistent correction for a set of shapes, Kern's I(i) (x) is calculated independently for each shape in the design, and the results are stored, without at first cutting off the negative going pieces. The values of I(i) (x) are then summed on a finite grid of points x(k), k = 1, 2, ....P, with each subsequent value being added to any already there. Thus, each shape is read only one time from s file, and its contribution to the resultant intensity I(x(k)) is calculated and added to the previously stored values of I(x(...