Method and apparatus for endpoint detection using partial least squares

• US 20050016947A1
• Filed: 03/25/2002
• Published: 01/27/2005
• Est. Priority Date: 03/23/2001
• Status: Active Grant

• Current Assignee: Tokyo Electron Limited
• Original Assignee: Tokyo Electron Limited
• Inventors: Yue, Hongyu, Fatke, David
• Granted Patent: US 7,297,287 B2
• Time in Patent Office: Days
• US Class Current: 216/2
• CPC Class Codes: H01J 37/32935 Monitoring and controlling tubes by information coming from the object and/or discharge
• CPC Class Codes: H01J 37/32963 End-point detection
• CPC Class Codes: H01L 22/20 Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps
• CPC Class Codes: H01L 2924/00 Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
• CPC Class Codes: H01L 2924/0002 Not covered by any one of groups H01L24/00, H01L24/00 and H01L2224/00

Abstract

An apparatus and method for detection of a feature etch completion within an etching reactor. The method includes determining a correlation matrix by recording first measured data regarding a first etch process over successive time intervals to form a first recorded data matrix, assembling a first endpoint signal matrix using target endpoint data for a specific etch process, performing a partial least squares analysis on the recorded data matrix and the first endpoint signal matrix to refine the recorded data matrix, and computing a correlation matrix based upon the refined recorded data matrix and the first endpoint signal matrix. The method further includes performing a second etch process to form a second recorded data matrix. The correlation matrix and the second recorded data matrix are analyzed to determine whether an endpoint of the second etch process has been achieved.

Claims

1. A method for detection of a feature etch completion, the method comprising the steps of:

determining a correlation matrix by;

recording first measured data regarding a first etch process over successive time intervals to form a first recorded data matrix,

assembling a first endpoint signal matrix using target endpoint data for a specific etch process,

performing a partial least squares analysis on the first recorded data matrix and the first endpoint signal matrix to refine the first recorded data matrix, and

computing a correlation matrix based upon the refined recorded data matrix and the first endpoint signal matrix; and

performing a second etch process to form a second recorded data matrix, wherein the correlation matrix and the second recorded data matrix are analyzed to determine whether an endpoint of the second etch process has been achieved.

2. The method according to claim 1, wherein said step of performing a partial least squares analysis includes the steps of:

calculating variable importance in projection data defined as an influence on the first endpoint signal matrix of the first measured data; and

refining the first recorded data matrix based upon an analysis of the variable importance in projection data.

3. The method according to claim 2, wherein the step of refining the first recorded data matrix comprises analyzing the variable importance in projection data to determine if a variable within the first recorded data matrix can be eliminated as having minimal impact on the first endpoint signal matrix.

4. The method according to claim 3, wherein the variable having a variable importance in projection data value below a predetermined threshold value is discarded.

5. The method according to claim 3, wherein the variable having a variable importance in projection data value within a predetermined range is discarded.

6. The method according to claim 3, wherein at least a first derivative of a variable importance in projection data value with respect to a variable number is used to select a threshold value for the variable importance in projection data below which the variable is discarded.

7. The method according to claim 1, wherein said step of performing a second etch process comprises the steps of:

initiating the second etch process within a processing chamber;

recording second measured data regarding the second etch process over successive time intervals to form the second recorded data matrix of at least one recorded data vector;

calculating at least one endpoint signal by multiplying the at least one recorded data vector and at least one weighting vector of the correlation matrix;

determining whether the endpoint has been achieved by inspecting the at least one end point signal; and

stopping the etch process when the endpoint has been achieved.

8. The method according to claim 1, wherein the first etch process and the second etch process are performed within a single processing chamber.

9. The method according to claim 1, wherein the correlation matrix is calculated for a selected etch process performed within a selected processing chamber.

10. The method according to claim 1, wherein the target data is determined by experimentation within a selected processing chamber, and wherein the selected processing chamber is utilized for the second etch process.

11. The method according to claim 1, wherein the measured data is optical emission data.

12. The method according to claim 1, wherein the measured data is electrical signal data.

13. The method according to claim 1, wherein the measured data is match network capacitor setting data.

14. The method according to claim 1, wherein the first recorded data matrix, the first endpoint signal matrix, and the correlation matrix are defined by a relationship:

{overscore (XB)}={overscore (Y)},

where {overscore (X)} represents the first recorded data matrix having m by n data points, {overscore (B)} represents the correlation matrix having n by p data points, and {overscore (Y)} represents the first endpoint signal matrix having m by p data points.

15. The method according to claim 1, wherein data of a given instant in time within the first recorded data matrix and the second recorded data matrix is mean-centered by computing a mean value of elements in a column of a respective matrix and subtracting the mean value from each element.

16. The method according to claim 1, wherein data of a given instant in time within the first recorded data matrix and the second recorded data matrix is normalized by determining a standard deviation of data in a column of a respective matrix.

17. An apparatus comprising:

an etching reactor configured to perform an etch process therein, said etch process being driven by a power source connected to said etching reactor;

an end-point detector for detecting an endpoint of said etching process, said end-point detector comprising a detecting section and a calculating section, said detecting section being configured to sequentially detect data relating to the etch process within said etching reactor, said calculating section being configured to determine a correlation matrix using first measured data regarding a first etch process over successive time intervals to form a first recorded data matrix, assemble a first endpoint signal matrix using target endpoint data for a specific etch process, perform a partial least squares analysis on the first recorded data matrix and the first endpoint signal matrix to refine the first recorded data matrix, compute a correlation matrix based upon the refined recorded data matrix and the first endpoint signal matrix, and form a second recorded data matrix for a second etch process, wherein said calculating section is configured to analyze the correlation matrix and the second recorded data matrix and produce an endpoint signal when an endpoint of the second etch-process has been achieved; and

a controller configured to receive said endpoint signal from said calculating section, said controller being configured to control said power source based upon said endpoint signal.

18. The apparatus according to claim 17, wherein said detecting section comprises a photodetector configured to sequentially detect an emission spectrum within said etching reactor.

19. The apparatus according to claim 18, wherein said photodetector section comprises a high resolution optical emission spectroscopy sensor.

20. The apparatus according to claim 18, wherein said etching reactor includes a vacuum chamber having a view window made of transparent material throughwhich said detecting section detects the emission spectrum.

21. The apparatus according to claim 17, wherein said etching reactor is a capacitively coupled plasma reactor including a vacuum chamber, a pair of parallel plate electrodes provided within said vacuum chamber, a gas injection line connected to said vacuum chamber, and a gas exhaust line connected to said vacuum chamber, wherein said power source is a high frequency power source connected to one of said pair of parallel plate electrodes.

22. The apparatus according to claim 17, wherein said etching reactor is selected from a group consisting essentially of a multi-frequency capacitively coupled plasma reactor, an inductively coupled plasma reactor, an electron cyclotron resonance reactor, and a helicon plasma reactor.

23. The apparatus according to claim 17, wherein said calculating section is configured to calculate variable importance in projection data defined as an influence on the first endpoint signal matrix of the first measured data, and refine the first recorded data matrix based upon an analysis of the variable importance in projection data.

24. The apparatus according to claim 23, wherein said calculating section is configured to refine the first recorded data matrix by analyzing the variable importance in projection data to determine if a variable within the first recorded data matrix can be eliminated as having minimal impact on the first endpoint signal matrix.

25. The apparatus according to claim 17, wherein said detecting section is configured to sequentially detect electrical signal data relating to the etch process within said etching reactor.

26. The apparatus according to claim 17, wherein said detecting section is configured to sequentially detect match network capacitor setting data relating to the etch process within said etching reactor.