Method to calibrate, predict, and control stochastic defects in EUV lithography

Patent metadata

Field	Value
Publication number	US-12561790-B2
Application number	US-202418894540-A
Country	US
Kind code	B2
Filing date	Sep 24, 2024
Priority date	Sep 26, 2023
Publication date	Feb 24, 2026
Grant date	Feb 24, 2026

Abstract

Using an initial probability of occurrence of a stochastic defect over an inspection area of a workpiece, one or more defects within the inspection area are imaged using an optical tool or an electron beam tool. A probability of occurrence of a stochastic defect at each of the defect locations is generated using the model. The defect locations are grouped into probability bins. A consistency between the initial probability and observed results is determined and the model can be tuned based on the consistency.

First claim

What is claimed is:

1 . A method comprising:
receiving, at a processor, an initial probability of occurrence of a stochastic defect over an inspection area of a workpiece, wherein the initial probability of occurrence of a stochastic defect is generated using a model; imaging one or more defects within the inspection area using an optical tool or an electron beam tool, wherein the defects are each at a defect location; generating, using the processor, a probability of occurrence of a stochastic defect at each of the defect locations using the model; determining a desired resolution of a probability prediction for each of the probability bins using the processor; grouping, using the processor, the defect locations into probability bins after determining the desired resolution; determining, using the processor, a consistency between the initial probability and observed results; and tuning the model based on the consistency.

2 . The method of claim 1 , further comprising determining an expected defect count within each of the probability bins using the processor.

3 . The method of claim 1 , wherein the consistency is determined using binary cross-entropy, RMSe of expected versus observed count, binomial test for significance, or a Brier score.

4 . The method of claim 1 , wherein the imaging uses the electron beam tool.

5 . The method of claim 4 , wherein the imaging occurs over a plurality of workpiece exposures.

6 . The method of claim 5 , further comprising determining a defect frequency of the defects based on a defect count.

7 . The method of claim 4 , wherein the defect locations are grouped by geometric pattern shapes on the workpiece.

8 . The method of claim 1 , further comprising generating the initial probability of occurrence of a stochastic defect with the model using the processor.

9 . A non-transitory computer readable medium storing a program configured to instruct the processor to execute the method of claim 1 .

10 . A system comprising:
an inspection tool configured to image a workpiece; and a processor in electronic communication with the inspection tool, wherein the processor is configured to:
receive an initial probability of occurrence of a stochastic defect over an inspection area of a workpiece, wherein the initial probability of occurrence of a stochastic defect is generated using a model;
send instructions to image one or more defects within the inspection area with an inspection using the inspection tool, wherein the defects are each at a defect location;
generate a probability of occurrence of a stochastic defect at each of the defect locations using the model;
determine a desired resolution of a probability prediction for each of the probability bins;
group the defect locations into probability bins after the desired resolution is determined;
determine a consistency between the initial probability and observed results; and
tune the model based on the consistency.

11 . The system of claim 10 , wherein the inspection tool is an optical tool or an electron beam tool.

12 . The system of claim 10 , wherein the processor is further configured to determine an expected defect count within each of the probability bins.

13 . The system of claim 10 , wherein the consistency is determined using binary cross-entropy, RMSe of expected versus observed count, binomial test for significance, or a Brier score.

14 . The system of claim 10 , wherein the inspection tool is an electron beam tool, and wherein the imaging occurs over a plurality of workpiece exposures.

15 . The system of claim 14 , wherein the processor is further configured to determine a defect frequency of the defects based on a defect count.

16 . The system of claim 14 , wherein the defect locations are grouped by geometric pattern shapes on the workpiece.

17 . The system of claim 10 , wherein the processor is further configured to generate the initial probability of occurrence of a stochastic defect with the model.

18 . A non-transitory computer-readable storage medium, comprising one or more programs for executing the following steps on one or more computing devices:
receive an initial probability of occurrence of a stochastic defect over an inspection area of a workpiece, wherein the initial probability of occurrence of a stochastic defect is generated using a model; and send instructions to image one or more defects within the inspection area with an inspection using an optical tool or an electron beam tool, wherein the defects are each at a defect location; generate a probability of occurrence of a stochastic defect at each of the defect locations using the model; determine a desired resolution of a probability prediction for each of the probability bins; group the defect locations into probability bins after the desired resolution is determined; determine a consistency between the initial probability and observed results; and tune the model based on the consistency.

Assignees

• Kla Corp

Inventors

• Vukkadala Pradeep
• Zhang Cao
• Burov Anatoly
• Parsey Guy
• Ko Kyeongeun
• Bakarian Sergei G
• Krek Janez
• Bai Kunlun
• Higgins Craig
• Graves John S
• Smith Mark D
• Biafore John J

Classifications

• G06T2207/30242

  Counting objects in image · CPC title

• G06T7/0006 Primary

  using a design-rule based approach · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/30148

  Semiconductor; IC; Wafer · CPC title

• G06T2207/30164

  Workpiece; Machine component · CPC title