Critical dimension uniformity enhancement techniques and apparatus

Patent metadata

Field	Value
Publication number	US-2016110858-A1
Application number	US-201514884670-A
Country	US
Kind code	A1
Filing date	Oct 15, 2015
Priority date	Oct 21, 2014
Publication date	Apr 21, 2016
Grant date	—

Abstract

Disclosed are methods and apparatus for inspecting a photolithographic reticle. Modeled images of a plurality of target features of the reticle are obtained based on a design database for fabricating the reticle. An inspection tool is used to obtain a plurality of actual images of the target features of the reticle. The modelled and actual images are binned into a plurality of bins based on image properties of the modelled and actual images, and at least some of the image properties are affected by one or more neighbor features of the target features on the reticle in a same manner. The modelled and actual images from at least one of the bins are analyzed to generate a feature characteristic uniformity map for the reticle.

First claim

What is claimed is:

1 . A method of inspecting a photolithographic reticle, the method comprising:
obtaining modeled images of a plurality of target features of the reticle based on a design database for fabricating the reticle; from an inspection tool, obtaining a plurality of actual images of the target features of the reticle; binning the modelled images and the actual images into a plurality of bins based on image properties of the modelled and actual images, wherein at least some of the image properties of each bin are affected by one or more neighbor features of the target features on the reticle in a same manner; and analyzing the modelled images and the actual images from at least one of the bins to generate a feature characteristic uniformity map for the reticle.

2 . The method of claim 1 , wherein the feature characteristic uniformity map for each bin is a critical dimension uniformity (CDU) map for each bin.

3 . The method of claim 2 , wherein each modelled image includes a target feature having a CD value and a surrounding area, and wherein each modelled image incorporates at least a portion of the optical properties of the inspection tool.

4 . The method of claim 3 , wherein each bin contains modeled images that contain a near-constant CD error for the CD values of the features of such bin, and the CD error variation of each bin is less than 30% of accuracy requirement on CDU measurement.

5 . The method of claim 3 , wherein each modelled and actual image may have a size that includes the corresponding feature image and an image area that is within a distance that is 10 times the point-spread-function of the inspection tool from such corresponding feature image.

6 . The method of claim 3 , wherein the image properties include one or more of the following characteristics: slope, shape, size, brightness, color, texture, moment of inertia, context, proximity or adjacency to other features, transparency/opaqueness, or resulting values from performing a Fourier transform or other image analysis technique on each image.

7 . The method of claim 6 , wherein the binning is accomplished by first sorting the modelled and actual images into a plurality of first bins based on one or more of the image properties that are not affected by one or more neighbor features of the target features and then sorting the modelled and actual images of each first bin into a plurality of second bins based on one or more of the image properties that are affected by one or more neighbor features of the target features.

8 . The method of claim 3 , wherein the binning includes using a principal component analysis (PCA) to determine which image properties to use in the binning.

9 . The method of claim 8 , wherein the binning includes weighting the images properties based on their relative optical effect.

10 . The method of claim 8 , wherein the binning includes using a locality-sensitive hashing process.

11 . The method of claim 8 , wherein the binning includes using a partitioning or agglomeration type clustering process.

12 . The method of claim 1 , further comprising:
analyzing the modelled images and the actual images from each of the bins to generate a feature characteristic uniformity map for each bin; analyzing at least a portion of the feature characteristic uniformity maps to determine whether the reticle is within specification; repairing or discarding the reticle if the reticle is determined to not be within specification; and using the reticle if it is determined to be within specification.

13 . The method of claim 12 , further comprising determining a root cause based on a particular one of the feature characteristic uniformity maps being out of specification and associated with a predefined signature.

14 . An inspection system for inspecting a photolithographic reticle, the system comprising:
illumination optics for generating and directing an incident beam towards the reticle; output optics for detecting actual images from the reticle in response to the incident beam; and at least one memory and at least one processor that are configured to initiate the following operations:
obtaining modeled images of a plurality of target features of the reticle based on a design database for fabricating the reticle;
using the inspection system to obtain a plurality of actual images of the target features of the reticle;
binning the modelled images and the actual images into a plurality of bins based on image properties of the modelled and actual images, wherein at least some of the image properties of each bin are affected by one or more neighbor features of the target features on the reticle in a same manner; and
analyzing the modelled images and the actual images from at least one of the bins to generate a feature characteristic uniformity map for the reticle.

15 . The system of claim 14 , wherein the feature characteristic uniformity map for each bin is a critical dimension uniformity (CDU) map for each bin.

16 . The system of claim 15 , wherein each modelled image includes a target feature having a CD value and a surrounding area, and wherein each modelled image incorporates at least a portion of the optical properties of the inspection tool.

17 . The system of claim 16 , wherein each bin contains modeled images that contain a near-constant CD error for the CD values of the features of such bin, and the CD error variation of each bin is less than 30% of accuracy requirement on CDU measurement.

18 . The system of claim 16 , wherein each modelled and actual image may have a size that includes the corresponding feature image and an image area that is within a distance that is 10 times the point-spread-function of the inspection tool from such corresponding feature image.

19 . The system of claim 16 , wherein the image properties include one or more of the following characteristics: slope, shape, size, brightness, color, texture, moment of inertia, context, proximity or adjacency to other features, transparency/opaqueness, or resulting values from performing a Fourier transform or other image analysis technique on each image

20 . The system of claim 19 , wherein the binning is accomplished by first sorting the modelled and actual images into a plurality of first bins based on one or more of the image properties that are not affected by one or more neighbor features of the target features and then sorting the modelled and actual images of each first bin into a plurality of second bins based on one or more of the image properties that are affected by one or more neighbor features of the target features.

21 . The system of claim 16 , wherein the binning includes using a principal component analysis (PCA) to determine which image properties to use in the binning.

22 . The system of claim 21 , wherein the binning includes weighting the images properties based on their relative optical effect.

23 . The system of claim 21 , wherein the binning includes using a locality-sensitive hashing process.

24 . The system of claim 21 , wherein the binning includes using a partitioning or agglomeration type clustering process.

25 . The system of claim 14 , further comprising:
analyzing the modelled images and the actual images from each of the bins to generate a feature characteristic uniformity map for each bin; analyzing at least a portion of the feature characteristic uniformity maps to deter

Assignees

• Kla Tencor Corp

Inventors

• Liu Yanwei
• Fang Hawren

Classifications

• G03F1/84 Primary

  Inspecting · CPC title

• G06V10/762

  using clustering, e.g. of similar faces in social networks · CPC title

• G06F18/23

  Clustering techniques · CPC title

• G06T2207/30148

  Semiconductor; IC; Wafer · CPC title

• G06T7/001

  using an image reference approach · CPC title