METHOD OF ENDPOINT DETECTION OF PLASMA ETCHING PROCESS USING MULTIVARIATE ANALYSIS

• US 20140106477A1
• Filed: 10/17/2013
• Published: 04/17/2014
• Est. Priority Date: 10/17/2012
• Status: Active Grant

• Current Assignee: Tokyo Electron Limited
• Original Assignee: Tokyo Electron Limited
• Inventors: CHEN, Yan, Komarov, Serguei, Vuong, Vi
• Granted Patent: US 9,330,990 B2
• Time in Patent Office: Days
• US Class Current: 438/9
• CPC Class Codes: G06F 17/16 Matrix or vector computation , e.g. matrix-matrix or matrix-vector multiplication, matrix factorization matrix transposition G06F7/78
• CPC Class Codes: H01J 37/32963 End-point detection
• CPC Class Codes: H01L 21/3065 Plasma etching; Reactive-ion etching
• CPC Class Codes: H01L 21/67069 for drying etching
• CPC Class Codes: H01L 22/26 Acting in response to an ongoing measurement without interruption of processing, e.g. endpoint detection, in-situ thickness measurement endpoint detection arrangements in CMP apparatus B24B37/013, in discharge apparatus H01J37/32

Abstract

Disclosed is a method for determining an endpoint of an etch process using optical emission spectroscopy (OES) data as an input. Optical emission spectroscopy (OES) data are acquired by a spectrometer attached to a plasma etch processing tool. The acquired time-evolving spectral data are first filtered and demeaned, and thereafter transformed into transformed spectral data, or trends, using multivariate analysis such as principal components analysis, in which previously calculated principal component weights are used to accomplish the transform. A functional form incorporating multiple trends may be used to more precisely determine the endpoint of an etch process. A method for calculating principal component weights prior to actual etching, based on OES data collected from previous etch processing, is disclosed, which method facilitates rapid calculation of trends and functional forms involving multiple trends, for efficient and accurate in-line determination of etch process endpoint.

Claims

1. A method for determining etch process endpoint data, comprising:

in a plasma etch processing tool, performing k plasma etch process runs, where k is an integer greater than zero, each of the k plasma etch process runs comprising steps of;

loading a substrate to be processed into the plasma etch processing tool, the plasma etch processing tool comprising a spectrometer having a detector comprising m pixels, each pixel corresponding to a different light wavelength;

igniting a plasma in the plasma etch processing tool;

collecting n optical emission spectroscopy (OES) data sets sampled at equal time intervals during each of k plasma etch process runs, each of the n optical emission spectroscopy (OES) data sets comprising m pixel intensities corresponding to m pixels of the spectrometer;

forming an n×

m optical emission spectroscopy (OES) data matrix [X] for each of k plasma etch process runs, each time sample occupying a row of the optical emission spectroscopy (OES) data matrix [X], the columns of the optical emission spectroscopy (OES) data matrix [X] corresponding to pixels of the spectrometer;

computing an n×

m average optical emission spectroscopy (OES) data matrix [X]^avg, wherein each element of the average optical emission spectroscopy (OES) data matrix [X]^avg is computed as an average of elements of optical emission spectroscopy (OES) data matrices [X] for the k etch process runs;

filtering noise from the average optical emission spectroscopy (OES) data matrix [X]^avg;

truncating each optical emission spectroscopy (OES) data matrix [X] to eliminate optical emission spectroscopy (OES) data acquired during plasma startup and for times beyond an etch process endpoint;

truncating the average optical emission spectroscopy (OES) data matrix [X]^avg to eliminate averaged optical emission spectroscopy (OES) data acquired during plasma startup and for times beyond the etch process endpoint;

calculating an n x m mean optical emission spectroscopy (OES) data matrix [S_avg], wherein each element of each of the columns of the mean optical emission spectroscopy (OES) data matrix [S_avg] is computed as an average of each of n pixel intensities of the average optical emission spectroscopy (OES) data matrix [X]^avg over the column of the average optical emission spectroscopy (OES) data matrix [X]^avg;

subtracting the mean optical emission spectroscopy (OES) data matrix [S_avg] from optical emission spectroscopy (OES) data matrices [X], to de-mean the optical emission spectroscopy (OES) data, and performing a principal component analysis [T]=([X]^[i]−

[S_avg])[P] on the de-meaned and non-normalized subtraction results, to obtain a transformed optical emission spectroscopy (OES) data vector [T] and principal component weights vector [P];

storing the mean optical emission spectroscopy (OES) data matrix [S_avg] for later use in in-situ determination of an etch process endpoint;

storing the principal component weights vector [P] for later use in in-situ determination of the etch process endpoint.

2. The method of claim 1, wherein the step of filtering noise comprises applying a moving average filter.

3. The method of claim 1, wherein the step of filtering noise comprises applying a filter whose characteristics are specific to the plasma etch processing tool.

4. The method of claim 1, further comprising:

selecting a functional form f(T_i) involving elements of the transformed optical emission spectroscopy (OES) data vector [T] as a trend variable for in-situ determination of the etch process endpoint.

5. The method of claim 4, wherein the selected functional form f(T_i) of the trend variable is a single element of the transformed optical emission spectroscopy (OES) data vector [T].

6. The method of claim 4, wherein the selected functional form f(T_i) of the trend variable contains a ratio of two elements of the transformed optical emission spectroscopy (OES) data vector [T].

7. The method of claim 4, further comprising:

computing a time evolution of at least one element of the transformed optical emission spectroscopy (OES) data vector [T];

determining at least one minimum value min(T_i) of the transformed optical emission spectroscopy (OES) data vector [T] from the computed time evolution of the at least one element of the transformed optical emission spectroscopy (OES) data vector [T];

storing the at least one determined minimum value min(T_i) of an element of the transformed optical emission spectroscopy (OES) data vector [T] for later use in in-situ determination of the etch process endpoint.

8. The method of claim 7, wherein the selected functional form f(T_i) of the trend variable is an element of the transformed optical emission spectroscopy (OES) data vector [T] from which the twice the minimum value min(_i) of the same element of the transformed optical emission spectroscopy (OES) data vector [T] has been subtracted.

9. The method of claim 5, wherein the selected functional form f(T_i) of the trend variable contains a ratio of two elements of the transformed optical emission spectroscopy (OES) data vector [T], wherein twice a minimum value min(T_i) of each of the elements of the transformed optical emission spectroscopy (OES) data vector [T] has been subtracted from the respective elements of the transformed optical emission spectroscopy (OES) data vector [T] prior to computing the ratio.

10. The method of claim 9, wherein the selected functional form f(T_i) of the trend variable involves a second and third element of the transformed optical emission spectroscopy (OES) data vector [T], and is of the form f(T_i)=(T₂−

2·

min(T₂))²/(T₃−

2·

min(T₃))².

11. The method for in-situ determination of etch process endpoint data, comprising:

providing a previously calculated and stored mean optical emission spectroscopy (OES) data matrix [S_avg];

providing a previously calculated and stored principal component weights vector [P];

loading a substrate into a plasma etch processing tool;

igniting a plasma in the plasma etch processing tool to initiate an etch process;

acquiring optical emission spectroscopy (OES) data sets from a spectrometer on the plasma etch processing tool at predetermined time intervals during the etch process;

from each acquired optical emission spectroscopy (OES) data set, subtracting the previously provided mean optical emission spectroscopy (OES) data matrix [S_avg], to de-mean each acquired optical emission spectroscopy (OES) data set;

transforming each acquired, de-meaned, and non-normalized optical emission spectroscopy (OES) data set into transformed optical emission spectroscopy (OES) data, by calculating at least one element of the transformed optical emission spectroscopy (OES) data vector [T] using the previously provided principal component weights vector [P];

from the calculated at least one element of the transformed optical emission spectroscopy (OES) data vector [T], further calculating a trend variable f(T_i);

detecting an endpoint of the etch process in-situ from the calculated values of the trend variable f(T_i) during the etch process.

12. The method of claim 11, further comprising:

stopping the etch process upon endpoint detection.

13. The method of claim 11, wherein a functional form of the trend variable f(T_i) is a calculated single element of the transformed optical emission spectroscopy (OES) data vector [T].

14. The method of claim 11, wherein a functional form of the trend variable f(T_i) contains a ratio of two calculated elements of the transformed optical emission spectroscopy (OES) data vector [T].

15. The method of claim 11, further comprising:

providing previously calculated and stored minimum value min(T_i) of at least one element of the transformed optical emission spectroscopy (OES) data vector [T].

16. The method of claim 15, wherein the functional form f(T_i) of the trend variable f(T_i) is an element of the transformed optical emission spectroscopy (OES) data vector [T] from which the twice the minimum value min(T_i) of the same element of the transformed optical emission spectroscopy (OES) data vector [T] has been subtracted.

17. The method of claim 15, wherein the selected functional form f(T_i) of the trend variable is a ratio of two elements of the transformed optical emission spectroscopy (OES) data vector [T], wherein twice a minimum value min(T_i) of each of the elements of the transformed optical emission spectroscopy (OES) data vector [T] has been subtracted from the respective elements of the transformed optical emission spectroscopy (OES) data vector [T] prior to computing the ratio.

18. The method of claim 17, wherein the selected functional form f(T_i) of the trend variable involves a second and third element of the transformed optical emission spectroscopy (OES) data vector [T], and is of the form f(T_i)=(T₂−

2·

min(T₂))²/(T₃−

2·

min(T₃))².

19. The method of claim 11, wherein the step of detecting an endpoint of the etch process comprises evaluating a differential of the trend variable f(T_i) during the etch process and detecting the endpoint of the etch process from the differentiated trend variable f(T_i).