Mask data design method

• US 6,243,855 B1
• Filed: 09/29/1998
• Issued: 06/05/2001
• Est. Priority Date: 09/30/1997
• Status: Expired due to Term

• Current Assignee: Toshiba Memory Corporation (Kioxia Corporation), Telefonaktiebolaget LM Ericsson
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Uno, Taiga, Hashimoto, Koji, Kobayashi, Sachiko, Yamamoto, Kazuko
• Primary Examiner(s): Smith, Matthew
• Assistant Examiner(s): THOMPSON, ANNETTE M
• Application Number:
  US09/161,959
• Time in Patent Office: 980 Days
• Field of Search: 716/19
• US Class Current: 716/53
• CPC Class Codes: G03F 1/36 Masks having proximity correction features; Preparation thereof, e.g. optical proximity correction [OPC] design processes
• CPC Class Codes: G03F 7/70441 Optical proximity correction [OPC]

Abstract

A correction target segment extracted from the design pattern is divided into lengths suited for correction. If the arrangement of the divided segments is a one-dimensional pattern, a correction value is obtained by conducting a one-dimensional process simulation to an arrangement within a predetermined distance from a divided segment in perpendicular direction. If the arrangement of the divided segments is a two-dimensional pattern, a correction value is obtained by two-dimensionally extracting a pattern included in a rectangular region having a predetermined distance from one point on the divided segment in perpendicular and horizontal directions and by conducting a two-dimensional process simulation to the extracted pattern.

Claims

1. A mask data design method correcting a design pattern and using the corrected design pattern as mask data to improve fidelity of a transferred pattern onto a wafer, said method comprising the steps of:

extracting a correction target segment from said design pattern;

dividing the extracted segment into lengths suited for correction;

determining whether characteristics of spatial arrangement of a divided segment comply to a predetermined one-dimensional rule; and

classifying, if the characteristics of spatial arrangement of the divided segment comply to the predetermined one-dimensional rule, said divided segment as a one-dimensional pattern and, if the characteristics of spatial arrangement of the divided segment does not comply to the predetermined one-dimensional rule, classifying said divided segment as a two-dimensional pattern; and

correcting said divided segment in accordance with classified pattern types.

2. A mask data design method according to claim 1, wherein the characteristics of the spatial arrangement of the divided segment are defined by dimension-characteristics of at least one of the divided segment, an adjacent segment located within a predetermined distance from the divided segment in a perpendicular direction, and a positional relationship between a segment of interest and the adjacent segment.

3. A mask data design method according to claim 1, wherein the step of correcting said divided segment includes the steps of:

obtaining, if the characteristics of the spatial arrangement of said divided segment is classified as a one-dimensional pattern, a correction value for said divided segment by conducting one of a one-dimensional process simulation, a one-dimensional pattern transfer test, and a combination of the one-dimensional process simulation and the one-dimensional pattern transfer test for a certain region including the divided segment; and

obtaining, if the characteristics of the spatial arrangement of said divided segment is classified as a two-dimensional pattern, a correction value for said divided segment by conducting one of a two-dimensional process simulation, a two-dimensional pattern transfer test, and a combination of the two-dimensional process simulation and the two-dimensional pattern transfer test for a certain region including the divided segment.

4. A mask data design method according to claim 3, wherein

the step of correcting said divided segment further includes the step of, while using a database storing characteristics of spatial arrangement of segments and correction values corresponding to the segments, taking out a correction value suited for said divided segment from the database and correcting said divided segment if the characteristics of the spatial arrangement of said divided segment is stored in the database.

5. A mask data design method according to claim 3, wherein

in the step of obtaining a correction value for said divided segment having the spatial arrangement classified as a one-dimensional pattern, the correction value is obtained for said divided segment by conducting one of the one-dimensional process simulation, the one-dimensional pattern transfer test and the combination of the one-dimensional process simulation and the one-dimensional pattern transfer test for a region including the divided segment and other segments located within a predetermined distance from the divided segment in perpendicular direction.

6. A mask data design method according to claim 3, wherein

in the step of obtaining a correction value for the divided segment having the spatial arrangement classified as a two-dimensional pattern, the correction value for said divided segment is obtained by extracting a pattern included in a rectangular region having a predetermined distance from one point on the divided segment in perpendicular and horizontal directions and by conducting one of the two-dimensional process simulation, the two-dimensional pattern transfer test and the combination of the two-dimensional process simulation and the two-dimensional pattern transfer test to the extracted pattern.

7. A mask data design method according to claim 3, wherein

in the step of obtaining a correction value for the divided segment having the spatial arrangement classified as a two-dimensional pattern, the correction value for said divided segment is obtained by two-dimensionally extracting a pattern included in a rectangular region having a predetermined distance from both terminations of the divided segment and from the divided segment and by conducting one of the two-dimensional process simulation, the two-dimensional pattern transfer test and a combination of the two-dimensional process simulation and the two-dimensional pattern transfer test to the extracted pattern.

8. A mask data design method according to claim 3, wherein one of the two-dimensional process simulation, the two dimensional pattern transfer test, and the combination of the two-dimensional process simulation and the two-dimensional pattern transfer test is conducted, and the thus obtained correction value is applied to a region adjacent to the divided segment classified as a two-dimensional pattern, even if the region includes the divided segment classified as a one-dimensional pattern.

9. A mask data design method according to claim 1, wherein in the step of correcting said divided segment, if a plurality of divided segments are densely located, a correction calculation is conducted for a region including the plurality of divided segments.

10. A mask data design method according to claim 1, wherein in the step of determining whether the characteristics of the spatial arrangement of the divided segment complies with a predetermined one-dimensional rule, it is determined that the arrangement of said divided segment can be approximated based on the one-dimensional rule when a length of the divided segment is more than a predetermined length.

11. A mask data design method according to claim 1, wherein in the step of determining whether the characteristics of the spatial arrangement of the divided segment complies with a predetermined one-dimensional rule, it is determined that the arrangement of said divided segment cannot be approximated to the one-dimensional rule when an adjacent segment within a distance R from a segment of interest in a perpendicular direction is located in neither perpendicular nor parallel direction of the segment of interest.

12. A mask data design method according to claim 1, wherein in the step of determining whether the characteristics of the spatial arrangement of the divided segment complies with a predetermined one-dimensional rule, it is determined that the arrangement of said divided segment cannot be approximated to the one-dimensional rule when a segment of interest is located in neither perpendicular nor parallel arrangement with respect to a mask data design coordinates axis.

13. A mask data design method according to claim 1, wherein in the step of determining whether the characteristics of the spatial arrangement of the divided segment complies with a predetermined one-dimensional rule, it is determined that the arrangement of said divided segment cannot be approximated to the one-dimensional rule when a pattern including a segment of interest is separated into rectangles and triangles wherein a ratio of a length to a width of the rectangle including the segment of interest falls within a predetermined range.

14. A mask data design method according to claim 1, wherein in the step of determining whether the characteristics of the spatial arrangement of the divided segment complies with a predetermined one-dimensional rule, it is determined that the arrangement of said divided segment cannot be approximated to the one-dimensional rule when a segment is in contact with a termination of a segment of interest with an angle more than a preset angle, with an extension of the segment of interest, and when a length of the segment contacting with the segment of interest is more than a predetermined length.

15. A mask data design method according to claim 1, wherein

in the step of dividing said segment, while observing within a predetermined distance from a noted segment in a perpendicular direction, the noted segment is divided such that a length of a segment after being divided is not shorter than a predetermined length based on an intersection between perpendicular lines drawn to the noted segment from topmost vertices of adjacent patterns opposite to a pattern including the noted segment or based on an adjacency of the intersection.

16. A mask data design method according to claim 1, wherein

in the step of dividing said segment, a noted segment is divided such that a length of a segment after being divided is not shorter than a predetermined length based on an intersection between perpendicular lines drawn to the noted segment from topmost vertices of a pattern, located within a predetermined range R in a direction perpendicular to the noted segment.

17. A mask data design apparatus for correcting a design pattern and using the corrected design pattern as mask data to improve fidelity of a transferred pattern to the design pattern, said apparatus comprising:

means for extracting a correction target segment from the design pattern;

means for dividing the extracted segment into lengths suited for correction;

means for classifying characteristics of spatial arrangement of divided segments into one-dimensional patterns and two-dimensional patterns depending on whether the characteristics of the spatial arrangement of the divided segments can be approximated to a one-dimensional rule; and

means for making corrections in accordance with types of classified segments.

18. A computer program product capable of being read by a computer and storing a program for creating mask data by correcting a design pattern to improve fidelity of a transferred pattern to the design pattern, said program executing the steps of:

extracting a correction target segment from the design pattern;

dividing the extracted segment into lengths suited for correction;

classifying characteristics of spatial arrangement of divided segments into one-dimensional patterns and two-dimensional patterns depending on whether the characteristics of the spatial arrangement of the divided segments can be approximated to a one-dimensional rule; and

making corrections in accordance with types of classified segments.