Method for automatic determination of semiconductor plasma chamber matching and source of fault by comprehensive plasma monitoring

• US 7,169,625 B2
• Filed: 07/25/2003
• Issued: 01/30/2007
• Est. Priority Date: 07/25/2003
• Status: Expired due to Fees

• Current Assignee: Applied Materials Incorporated
• Original Assignee: Applied Materials Incorporated
• Inventors: Lian, Lei, Davis, Matthew F.
• Primary Examiner(s): Lebentritt; Michael
• Assistant Examiner(s): STEVENSON, ANDRE C
• Application Number:
  US10/628,001
• Publication Number:
  US 20050019961A1
• Time in Patent Office: 1,285 Days
• Field of Search: 438/5, 438/7, 438 10- 11, 438 16- 18, 438 22- 24, 438/29, 438/31, 438 34- 36, 438128-130, 438/149, 438/484, 438/538, 438/14
• US Class Current: 438/9
• CPC Class Codes: G01N 21/73 using plasma burners or torches
• CPC Class Codes: H01J 37/32174 Circuits specially adapted for controlling the RF discharge
• CPC Class Codes: H01J 37/32935 Monitoring and controlling tubes by information coming from the object and/or discharge
• CPC Class Codes: H01J 37/32972 Spectral analysis

Abstract

A method and apparatus for automatic determination of semiconductor plasma chamber matching a source of fault are provided. Correlated plasma attributes are measured for process used for calibration both in a chamber under study and in a reference chamber. Principal component analysis then is performed on the measured correlated attributes so as to generate steady principal components and transitional principal components; and these principal components are compared to reference principal components associated with a reference chamber. The process used for calibration includes a regular plasma process followed by a process perturbation of one process parameter. Similar process perturbation runs are conducted several times to include different perturbation parameters. By performing inner products of the principal components of chamber under study and the reference chamber, matching scores can be reached. Automatic chamber matching can be determined by comparing these scores with preset control limits. The potential source(s) of chamber fault can also be identified by the lowest matching score(s).

Claims

1. A method of evaluating a process chamber, comprising:

performing a first substrate processing operation and a first process perturbation operation in a reference chamber;

collecting optical emission spectroscopy (OES) data and radio frequency (RF) data during the first substrate processing operation and the first process perturbation operation;

performing a first multivariate analysis on the collected OES and RF data from the reference chamber to produce first multivariate analysis results;

performing a second substrate processing operation and a second process perturbation operation in a process chamber that is under study, where the first process operation and the second process operation are similar, and the first process perturbation operation and the second process perturbation operation are similar;

collecting OES data and RF data during the second process operation and the second process perturbation operation;

performing a second multivariate analysis on the collected OES and RF data from the chamber under study to produce second multivariate analysis results; and

comparing the second multivariate analysis results from the chamber under study to the first multivariate analysis results from the reference chamber.

2. The method of claim 1 further comprising using a result of comparing the second multivariate analysis results to the first multivariate analysis results to calibrate the chamber under study.

3. The method of claim 1, further comprising:

using a result of comparing the second multivariate analysis results to the first multivariate analysis results to identify a fault in the chamber under study.

4. The method of claim 1, wherein the first and second multivariate analyses comprise principal component analysis (PCA).

5. The method of claim 1, further comprising:

determining if the OES and/or RF data collected from the reference chamber need to be manipulated;

performing data manipulation on the data collected from the reference chamber;

performing a multivariate analysis on the revised OES and RF data from the reference chamber; and

manipulating the OES and/or RF data collected from the process chamber that is under study according to the data manipulation scheme developed for the reference chamber.

6. The method of claim 5, wherein steady principal components and transitional principal components are identified for the reference chamber and the chamber under study by principal component analysis.

7. The method of claim 5, wherein comparing the results from the first and second multivariate analyses further comprises performing an inner product of the identified principal components for the reference chamber and the chamber under study to generate matching scores.

8. The method of claim 5, further comprising:

using decomposition of OES and RF data matrixes to yield scores for principal components to determine if the OES and/or RF data need to be manipulated.

9. The method of claim 5, further comprising:

enhancing weak signals either by amplifying post perturbation signals, or by selecting a narrower and more sensitive wavelength range or frequency range.

10. The method of claim 1, further comprising:

comparing the matching scores to a pre-established matching control limit.

11. The method of claim 1, wherein an amount of process parameter perturbation is less than 50%.

12. The method of claim 1, wherein the process chamber is an etching chamber.

13. The method of claim 1, wherein the process chamber is a deposition chamber.

14. The method of claim 3, further comprising:

accessing a library of stored diagnosis information; and

searching the diagnosis information to determine a solution to repair the identified fault.

15. A method of evaluating a process chamber, comprising:

performing a first substrate processing operation, followed by a first process perturbation operation of the first substrate process operation in a reference process chamber;

collecting data of one or more plasma attributes during the substrate processing operation and the following first process perturbation operation in the reference process chamber;

determining if the collected data of plasma attributes of the first process operation and the first process perturbation operation in the reference process chamber need to be manipulated;

performing data manipulation on the data of plasma attributes of the reference chamber that are identified to need data manipulation;

performing a multivariate analysis on the revised data of plasma attributes of the reference process chamber to produce first multivariate analysis results;

performing a second substrate processing operation, followed by a second process perturbation operation of the first substrate process operation in a process chamber that is under study, wherein the first process operation and the second process operation are similar, and the first process perturbation operation and the second process perturbation operation are similar;

collecting data of one or more plasma attributes during the second substrate processing operation and the following second process perturbation operation in the process chamber that is under study;

performing data manipulation on the data of plasma attributes of the process chamber under study according to the data manipulation scheme used by the reference chamber;

performing a multivariate analysis on the revised data of plasma attributes of the process chamber under study to produce second multivariate analysis results;

comparing the second multivariate analysis results from the chamber under study to the first multivariate analysis results from the reference chamber.

16. The method of claim 15, wherein the one or more plasma attributes are selected from the group consisting of optical electromagnetic emission, RF power, wafer reflectance, process pressure, process temperature, and the combinations thereof.

17. The method of claim 15, wherein the one or more plasma attributes are optical electromagnetic emission, RF power, and wafer reflectance.

18. The method of claim 15, wherein the plasma attribute is wafer reflectance.

19. The method of claim 15 further comprising using a result of comparing the second multivariate analysis results to the first multivariate analysis results to calibrate the chamber under study.

20. The method of claim 15, further comprising:

using a result of comparing the second multivariate analysis results to the first multivariate analysis results to identify a fault in the chamber under study.

21. The method of claim 15, wherein the first and second multivariate analyses comprise principal component analysis (PCA).

22. The method of claim 21, wherein steady principal components and transitional principal components are identified for the reference chamber and the chamber under study by principal component analysis.

23. The method of claim 15, wherein comparing the results from the first and second multivariate analyses further comprises performing an inner product of the identified principal components for the reference chamber and the chamber under study to generate matching scores.

24. The method of claim 15, wherein determining if the collected data of plasma attributes need to be manipulated further comprising:

using decomposition of plasma attributes data matrixes for the reference chamber to yield scores for principal components to determine if the data of plasma attributes of the reference chamber need to be manipulated.

25. The method of claim 15, wherein the performing data manipulation further comprising:

enhancing weak signals either by amplification, or by selecting a narrower and more sensitive wavelength range or frequency range.

26. The method of claim 23, further comprising:

comparing the matching scores to a pre-established matching control limit.

27. The method of claim 15, wherein the first and second process perturbations are performed by varying at least one process parameters.

28. The method of claim 27, wherein the at least one process parameter is varied by an amount less than 50% of the un-perturbed process parameter.

29. The method of claim 15, wherein the process chamber is an etching chamber.

30. The method of claim 15, wherein the process chamber is a deposition chamber.

31. The method of claim 20, further comprising:

accessing a library of stored diagnosis information; and

searching the diagnosis information to determine a solution to repair the identified fault.