Endpoint detection algorithm for atomic layer etching (ALE)

• US 10,453,653 B2
• Filed: 03/08/2017
• Issued: 10/22/2019
• Est. Priority Date: 09/02/2016
• Status: Active Grant

• Current Assignee: Tokyo Electron Limited
• Original Assignee: Tokyo Electron Limited
• Inventors: Chen, Yan, Tian, Xinkang, Ferns, Jason
• Primary Examiner(s): Duclair, Stephanie P
• Application Number:
  US15/453,555
• Publication Number:
  US 20180068831A1
• Time in Patent Office: 958 Days
• Field of Search: None
• CPC Class Codes: H01J 2237/334 Etching
• CPC Class Codes: H01J 37/32009 Arrangements for generation of plasma specially adapted for examination or treatment of objects, e.g. plasma sources plasma generation in general H05H1/24
• CPC Class Codes: H01J 37/32917 Plasma diagnostics
• CPC Class Codes: H01J 37/32963 End-point detection
• CPC Class Codes: H01J 37/32972 Spectral analysis
• CPC Class Codes: H01L 21/3065 Plasma etching; Reactive-ion etching
• CPC Class Codes: H01L 21/30655 comprising alternated and repeated etching and passivation steps, e.g. Bosch process
• 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

Described herein are architectures, platforms and methods for determining endpoints of an optical emission spectroscopy (OES) data acquired from a plasma processing system. The OES data, for example, includes an absorption—step process, a desorption—step process, or a combination thereof. In this example, the OES data undergoes signal synchronization and transient signal filtering prior to endpoint determination, which may be implemented through an application of a moving average filter.

Claims

1. A method for determining process endpoint data in a plasma processing system, comprising:

receiving optical emission spectroscopy (OES) data from a plasma processing chamber of the plasma processing system that performs a multi-step plasma process of a substrate that is an atomic layer etch (ALE) process, wherein the multi-step plasma process is cyclical and comprises at least two process steps that are cyclically repeated, and the at least two process steps comprise an absorption step and a desorption step;

determining a first quiescent portion of the received OES data, the first quiescent portion is an average of the absorption step of the ALE process;

determining a second quiescent portion of the received OES data, the second quiescent portion is an average of a desorption step of the ALE process; and

determining a first endpoint of the determined first quiescent portion.

2. The method of claim 1, wherein the at least two process steps includes a transient signal—

filtered absorption step or a transient signal—

filtered desorption step.

3. The method of claim 1, wherein the receiving of optical emission spectroscopy (OES) data is synchronized with the at least two process steps that are cyclically repeated.

4. The method of claim 1, further comprising:

removing transients associated with switching the at least two process steps prior to the determining of the first quiescent portion.

5. The method of claim 1, wherein the first quiescent portion is derived from an upper envelope of the OES data acquired.

6. The method of claim 1, further comprising:

determining an average of the first quiescent portion and the second quiescent portion.

7. The method of claim 6, wherein the determining of the first endpoint is based from the determined second quiescent portion.

8. The method of claim 6, wherein the determining of the first endpoint is based from the average of the first quiescent portion and the second quiescent portion.

9. The method of claim 1, wherein the determining the first quiescent portion comprises an application of a moving average filter to the at least two process steps.

10. The method of claim 1, further comprising:

removing transient signals of the received OES data prior to the determining of the first endpoint based on a second quiescent signal.

11. The method of claim 10, wherein determining a second endpoint based on the second quiescent data comprises a principal component analysis.

12. The method of claim 1, wherein determining of the first endpoint comprises a principal component analysis.

13. A method for determining process endpoint data in a plasma processing system, comprising:

receiving optical emission spectroscopy (OES) data from a plasma processing chamber of the plasma processing system that performs a multi-step plasma process of a substrate, wherein the multi-step plasma process is cyclical and comprises at least two process steps that are cyclically repeated;

determining a first quiescent portion of the received OES data, the first quiescent portion is an average of a filtered process-step of the multi-step plasma process; and

determining a first endpoint of the determined first quiescent portion,

wherein determining the first endpoint comprises;

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];

forming optical emission spectroscopy (OES) data sets from a first quiescent signal at predetermined time intervals during the cyclical multi-step plasma process;

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

transforming each 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 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); and

detecting the first endpoint of the cyclical multi-step plasma process from the calculated values of the trend variable f(T_i) during the cyclical multi-step plasma process.