Reticle inspection using near-field recovery

Patent metadata

Field	Value
Publication number	US-9478019-B2
Application number	US-201514702336-A
Country	US
Kind code	B2
Filing date	May 1, 2015
Priority date	May 6, 2014
Publication date	Oct 25, 2016
Grant date	Oct 25, 2016

Abstract

Systems and methods for detecting defects on a reticle are provided. The embodiments include generating and/or using a data structure that includes pairs of predetermined segments of a reticle pattern and corresponding near-field data. The near-field data for the predetermined segments may be determined by regression based on actual image(s) of a reticle generated by a detector of a reticle inspection system. Inspecting a reticle may then include separately comparing two or more segments of a pattern included in an inspection area on the reticle to the predetermined segments and assigning near-field data to at least one of the segments based on the predetermined segment to which it is most similar. The assigned near-field data can then be used to simulate an image that would be formed for the reticle by the detector, which can be compared to an actual image generated by the detector for defect detection.

First claim

What is claimed is:

1. A computer-implemented method for detecting defects on a reticle, comprising:
separating a pattern included in an inspection area on the reticle into two or more segments;
separately comparing the two or more segments to predetermined segments included in a data structure, wherein the data structure comprises pairs of the predetermined segments of a reticle pattern and corresponding near-field data;
assigning near-field data to at least one of the two or more segments based on one of the predetermined segments to which the at least one of the two or more segments is most similar;
generating near-field data for the inspection area based on the assigned near-field data;
simulating an image, based on the generated near-field data, of the inspection area that would be formed by a detector of a reticle inspection system;
acquiring an actual image of the inspection area on a physical version of the reticle generated by the detector; and
detecting defects on the reticle by comparing the simulated image to the actual image, wherein said separating, separately comparing, assigning, generating, simulating, acquiring, and detecting are performed with one or more computer systems.

2. The method of claim 1 , further comprising determining the pairs of the predetermined segments and the corresponding near-field data by separating pattern information for the reticle or another reticle into the predetermined segments and determining the near-field data for the predetermined segments based on one or more actual images of the reticle or the other reticle acquired by the detector or another detector.

3. The method of claim 2 , wherein the pattern information is acquired from design data for the reticle.

4. The method of claim 2 , further comprising selecting the predetermined segments from the pattern information based on number of instances of the predetermined segments in an entirety of the pattern for the reticle.

5. The method of claim 2 , wherein the one or more actual images of the reticle or the other reticle are acquired at more than one focus setting by the detector or the other detector.

6. The method of claim 2 , wherein the near-field data for the predetermined segments is determined by regression based on the one or more actual images of the reticle or the other reticle.

7. The method of claim 6 , wherein the regression comprises a Hopkins phase approximation.

8. The method of claim 6 , wherein the regression does not comprise thin-mask approximations.

9. The method of claim 6 , wherein the regression does not comprise Kirchhoff approximations.

10. The method of claim 1 , further comprising determining the pairs of the predetermined segments and the corresponding near-field data by separating pattern information for the reticle or another reticle into the predetermined segments and determining the near-field data for the predetermined segments by numerically solving equations that govern electromagnetic fields for the predetermined segments of the reticle pattern.

11. The method of claim 10 , wherein the pattern information is acquired from design data for the reticle.

12. The method of claim 10 , further comprising selecting the predetermined segments from the pattern information based on number of instances of the predetermined segments in an entirety of the pattern for the reticle.

13. The method of claim 1 , wherein diameters of the two or more segments are larger than an optical proximity distance for the reticle.

14. The method of claim 1 , wherein said separately comparing comprises comparing gray scale images for the two or more segments to gray scale images for the predetermined segments.

15. The method of claim 1 , wherein said separately comparing comprises comparing coordinates of vertices of polygons in one of the two or more segments to coordinates of vertices of polygons in the predetermined segments, and wherein a solid-body translation of the coordinates of the vertices in the one of the two or more segments and the predetermined segments is not considered a difference during said separately comparing.

16. The method of claim 1 , wherein when the one of the predetermined segments to which the at least one of the two or more segments is most similar is not an exact match for the at least one of the two or more segments, assigning the near-field data to the at least one of the two or more segments comprises determining a mapping of the one of the predetermined segments to the at least one of the two or more segments, applying the mapping to the near-field data corresponding to the one of the predetermined segments to generate modified near-field data, and assigning the modified near-field data to the at least one of the two or more segments.

17. The method of claim 1 , wherein when one of the predetermined segments to which at least one other of the two or more segments is most similar cannot be found for the at least one other of the two or more segments, the method further comprises determining near-field data for the at least one other of the two or more segments based on the actual image generated by the detector and adding the determined near-field data to the data structure.

18. The method of claim 1 , further comprising simulating one or more characteristics of patterned features that would be formed on a wafer in a process performed with the reticle by inputting the generated near-field data into a model, measuring the one or more characteristics of the patterned features that have been formed on the wafer in the process, comparing the simulated one or more characteristics to the measured one or more characteristics, and altering one or more parameters of the model based on results of said comparing.

19. The method of claim 18 , further comprising, for at least one of the defects detected on the reticle, determining near-field data for the at least one of the defects based on the actual image of the at least one of the defects and simulating how the at least one of the defects would be printed on the wafer in the process by inputting the determined near-field data for the at least one of the defects into the model.

20. A computer-implemented method for setting up a reticle inspection process, comprising:
separating pattern information for a reticle into predetermined segments;
determining near-field data for the predetermined segments based on one or more actual images of the reticle acquired by a detector of a reticle inspection system, wherein the near-field data for the predetermined segments is determined by regression based on the one or more actual images of the reticle;
generating a data structure that comprises pairs of the predetermined segments and the near-field data corresponding to the predetermined segments; and
setting up the reticle inspection process by incorporating information for the generated data structure in a recipe for the reticle inspection process such that during the reticle inspection process, two or more segments in an inspection area on the reticle or another reticle are compared to the predetermined segments and the near-field data for one of the predetermined segments to which at least one of the two or more segments is most similar is assigned to the at least one of the two or more segments, wherein the separating, determining, generating, and setting up steps are performed by one or more computer systems.

21. A computer-implemented method for setting up a reticle inspection process, comprising:
separating pattern information f

Assignees

• Kla Tencor Corp
• Kla Tencor Corp

Inventors

• Sezginer Abdurrahman
• Shi Rui-Fang

Classifications

• G01N2021/95676

  Masks, reticles, shadow masks · CPC title

• G01N21/9501

  Semiconductor wafers (manufacturing processes per se of semiconductor devices implementing a measuring step H10P74/20) · CPC title

• G06T7/001

  using an image reference approach · CPC title

• G01N21/956 Primary

  Inspecting patterns on the surface of objects {(contactless testing of electronic circuits G01R31/308; testing currency G07D; manufacturing processes per se of semiconductor devices implementing a measuring step H10P74/20)} · CPC title

• G01N2201/10

  Scanning · CPC title