Apparatus and method for segmenting edges for optical proximity correction

1. A computer implemented method for modifying a mask data set, comprising:

calculating a mathematical derivative of a figure-of-merit with respect to a mask function, wherein the figure-of-merit is indicative of a data set defined by a plurality of polygon edges and the mathematical derivative of the figure of merit is derived from a computational model of a patterning process; and

utilizing a computer to segment polygon edges in response to said step of calculating.

2. The method of modifying a mask data set according to claim 1 , further comprising:

moving at least one segment of said polygon edges in response to said step of segmenting.

3. The method of modifying a mask data set according to claim 2 , wherein the steps of segmenting and moving are repeated until said figure-of-merit ceases to significantly change.

4. The method of modifying a mask data set according to claim 2 , wherein the steps of segmenting and moving are repeated until a predetermined computational time is reached.

5. The method of modifying a mask data set according to claim 2 , wherein the steps of segmenting and moving are repeated a predetermined number of times.

6. The method of modifying a mask data set according to claim 1 , wherein said step of segmenting includes: segmenting said polygon edges where said mathematical derivative changes its sign.

7. The method of modifying a mask data set according to claim 1 , further comprising:

applying optical proximity correction to edge segments created in the step of segmenting.

8. A computer implemented method for modifying a lithography data set, comprising:

assigning a figure-of-merit to any data set defined by a plurality of polygons, each polygon having edges;

calculating a mathematical derivative of said figure-of-merit of said data set relative to a displacement of at least one infinitesimal section of an edge; and

segmenting said edge relative to at least one point at which said mathematical derivative changes its sign; and

utilizing a computer to move at least one said edge segment in a way that increases said figure-of-merit to form an improved set of polygons.

9. A method of modifying a mask data set according to claim 8 , further comprising:

increasing the figure-of-merit by moving at least one edge prior to the step of segmenting.

10. A method of modifying a mask data set according to claim 8 , further comprising:

repeating said steps of segmenting and moving iteratively.

11. The method of modifying a lithography data set according to claim 8 , wherein said set of improved polygons are used to form a pattern on a photomask.

12. The method of modifying a lithography data set according to claim 8 , wherein said set of improved polygons are used in a mask-less lithography data set.

13. The method of modifying a mask data set according to claim 8 , wherein said figure-of-merit is indicative of how well said data set meets the design goals of a proposed integrated circuit.

14. The method of modifying a mask data set according to claim 13 , wherein said figure of merit is derived from a computational model of a patterning process.

15. The method of modifying a mask data set according to claim 14 , wherein said computational model substantially predicts at least one of the following effects:

proximity effect and fogging in mask writing;

mask etch;

optical imaging;

photoresist blur;

resist develop; and

wafer etch.

16. The method of modifying a mask data set according to claim 14 , wherein said step of assigning a figure-of-merit to a data set defined by a plurality of polygon edges includes:

calculating a measure of difference between a pattern predicted by said computational model and a desired target pattern.

17. The method of modifying a mask data set according to claim 16 , wherein said step of calculating is performed for more than one dose value.

18. The method of modifying a mask data set according to claim 16 , wherein said step of calculating is performed for more than one defocus value.

19. The method of modifying a mask data set according to claim 16 , wherein said step of calculating is performed for more than one dose and more than one defocus value.

20. The method of modifying a mask data set according to claim 16 , wherein calculating said measure of difference comprises evaluating differences between said target pattern and said pattern predicted by said computational model, at a plurality of target points.

21. The method of modifying a mask data set according to claim 20 , wherein said measure of difference is related to a sum of squares of said differences.

22. The method of modifying a mask data set according to claim 20 , wherein said measure of difference is related to a sum of absolute values of said differences.

23. The method of modifying a mask data set according to claim 20 , wherein said measure of difference is related to the greatest of absolute values of said differences.

24. The method of modifying a mask data set according to claim 20 , wherein said measure of difference is a continuous function of movement of said polygon edges.

25. The method of modifying a mask data set according to claim 14 , wherein said step of assigning a figure-of-merit includes a combination of two or more of the following:

calculating a measure of difference between a pattern predicted by said computational model and a desired target pattern;

calculating a difference in the pattern predicted by said computational model wherein the difference is in response to a change in the exposure dose;

calculating a difference in the pattern predicted by said computational model wherein the difference is in response to a change in the defocus.

26. The method of modifying a mask data set according to claim 14 , wherein said step of assigning a figure-of-merit to a data set includes:

calculating a measure of image contrast.

27. The method of modifying a mask data set according to claim 14 , wherein said step of assigning a figure-of-merit to a data set includes:

calculating a measure of image slope.

28. The method of modifying a mask data set according to claim 14 , wherein said step of assigning a figure-of-merit includes:

calculating a difference in the pattern predicted by said computational model; wherein the difference is in response to a change in the exposure dose.

29. The method of modifying a mask data set according to claim 14 , wherein said step of assigning a figure-of-merit includes:

calculating a difference in the pattern predicted by said computational model; wherein the difference is in response to a change in the defocus.

30. A method of modifying a mask data set according to claim 8 , wherein said step of segmenting at least one polygon edge in said plurality of polygon edges includes:

calculating a derivative of said figure-of-merit of said data set relative to the displacement of at least one infinitesimal section of an edge; and

segmenting said edge relative to at least one point at which said derivative changes its sign.

31. The method of modifying a mask data set according to claim 30 , wherein said derivative is calculated explicitly.

32. A computer implemented method for correcting photolithography proximity, comprising:

adjusting a data set indicative of a plurality of polygon edges defined by a figure-of-merit which provides a measure of how well the data set meets the design goals for an integrated circuit layout;

segmenting one or more of the individual polygon edges at a point where a mathematical derivative of the figure-of-merit changes sign, wherein the mathematical derivative of the figure of merit is derived from a computational model of a patterning process;

moving the segmented edges until a set of optimized polygon edges results and

utilizing a computer to perform an optical proximity correction on said set of optimized polygon edges in order to correct for inherent distortions resulting from a photolithography process.