Method to predict yield of a device manufacturing process

Patent metadata

Field	Value
Publication number	US-11714357-B2
Application number	US-202117363057-A
Country	US
Kind code	B2
Filing date	Jun 30, 2021
Priority date	May 5, 2017
Publication date	Aug 1, 2023
Grant date	Aug 1, 2023

Abstract

A method and associated computer program for predicting an electrical characteristic of a substrate subject to a process. The method includes determining a sensitivity of the electrical characteristic to a process characteristic, based on analysis of electrical metrology data including electrical characteristic measurements from previously processed substrates and of process metrology data including measurements of at least one parameter related to the process characteristic measured from the previously processed substrates; obtaining process metrology data related to the substrate describing the at least one parameter; and predicting the electrical characteristic of the substrate based on the sensitivity and the process metrology data.

First claim

The invention claimed is:

1. A method comprising:
obtaining a machine learning model trained to metrology data comprising measurements of an electrical characteristic from previously processed substrates and process metrology data comprising measurements of at least one parameter related to a process characteristic measured from the previously processed substrates;
obtaining process metrology data related to a substrate comprising the at least one parameter; and
providing the obtained process metrology data to the machine learning model for predicting an electrical characteristic of the substrate; and
dynamically updating the machine learning model based on new process metrology data and/or new metrology data.

2. The method according to claim 1 , wherein the obtained process metrology data comprises a value and/or an uncertainty of the at least one parameter at a plurality of locations on the substrate.

3. The method according to claim 1 , wherein the machine learning model comprises a logistic model.

4. The method according to claim 1 , wherein the machine learning model is based on a method of reinforcement learning.

5. The method according to claim 1 , wherein the metrology data has been obtained, at least in part, using voltage contrast metrology or electrical probe measurements.

6. The method according to claim 1 , wherein the process metrology data comprises simulated data.

7. A method comprising:
obtaining, across a substrate, values of a control parameter for a process involving lithographic processing of substrates;
obtaining values of a yield parameter across the substrate;
correlating the values of the yield parameter to the values of the control parameter to obtain a model relating control parameter values to expected yield parameter values; and
determining the control parameter based on the model and an expected range of the control parameter associated with the process, wherein the determining the control parameter comprises determining a metrology offset between a nominal optimal control parameter value as measured by a metrology device and an actual optimal control parameter value which improves or optimizes yield.

8. The method according to claim 7 , wherein the control parameter is overlay and an actual optimal control parameter value is a non-zero value.

9. The method according to claim 7 , further comprising controlling the process for subsequent substrates by driving the control parameter towards an optimal control parameter value.

10. The method according to claim 7 , wherein the values of the yield parameter have been obtained, at least in part, using voltage contrast metrology or electrical probe measurements.

11. A computer program product comprising a non-transitory computer-readable medium comprising instructions therein, the instructions, upon execution by a processor system, configured to cause the processor system to at least:
obtain, across a substrate, values of a control parameter for a process involving lithographic processing of substrates;
obtain values of a yield parameter across the substrate;
correlate the values of the yield parameter to the values of the control parameter to obtain a model relating control parameter values to expected yield parameter values; and
determine the control parameter based on the model and an expected range of the control parameter associated with the process, wherein the determination of the control parameter comprises determination of a metrology offset between a nominal optimal control parameter value as measured by a metrology device and an actual optimal control parameter value which improves or optimizes yield.

12. The computer program product according to claim 11 , wherein the control parameter is overlay and an actual optimal control parameter value is a non-zero value.

13. The computer program product according to claim 11 , wherein the instructions are further configured to cause the processor system to control the process for subsequent substrates by driving the control parameter towards an optimal control parameter value.

14. The computer program product according to claim 11 , wherein the values of the yield parameter have been obtained, at least in part, using voltage contrast metrology or electrical probe measurements.

15. A computer program product comprising a non-transitory computer-readable medium comprising instructions therein, the instructions, upon execution by a processor system, configured to cause the processor system to at least:
obtain a machine learning model trained to metrology data comprising measurements of an electrical characteristic from previously processed substrates and process metrology data comprising measurements of at least one parameter related to a process characteristic measured from the previously processed substrates;
obtain process metrology data related to a substrate comprising the at least one parameter;
provide the obtained process metrology data to the machine learning model for predicting an electrical characteristic of the substrate; and
dynamically update the machine learning model based on new process metrology data and/or new metrology data.

16. The computer program product according to claim 15 , wherein the obtained process metrology data comprises a value and/or an uncertainty of the at least one parameter at a plurality of locations on the substrate.

17. The computer program product according to claim 15 , wherein the machine learning model comprises a logistic model.

18. The computer program product according to claim 15 , wherein the machine learning model is based on a method of reinforcement learning.

19. The computer program product according to claim 15 , wherein the metrology data has been obtained, at least in part, using voltage contrast metrology or electrical probe measurements.

20. The computer program product according to claim 15 , wherein the process metrology data comprises simulated data.

Assignees

• Asml Netherlands Bv

Inventors

• Ypma Alexander
• Tabery Cyrus Emil
• Van Gorp Simon Hendrik Celine
• Lin Chenxi
• Sonntag Dag
• Cekli Hakki Ergün
• Alvarez Sanchez Ruben
• Liu Shih-Chin
• Hastings Simon Philip Spencer
• Menchtchikov Boris
• De Ruiter Christiaan Theodoor
• Ten Berge Peter
• Lercel Michael James
• Duan Wei
• Guittet Pierre-Yves Jerome Yvan

Classifications

• G03F7/706837

  Data analysis, e.g. filtering, weighting, flyer removal, fingerprints or root cause analysis · CPC title

• G03F7/70491 Primary

  Information management, e.g. software; Active and passive control, e.g. details of controlling exposure processes or exposure tool monitoring processes · CPC title

• G03F7/705 Primary

  Modelling or simulating from physical phenomena up to complete wafer processes or whole workflow in wafer productions · CPC title

• G03F7/70658 Primary

  Electrical testing · CPC title

• G03F7/706835 Primary

  Metrology information management or control · CPC title