Applying dimensional reduction to spectral data from polishing substrates

Patent metadata

Field	Value
Publication number	US-9833874-B2
Application number	US-201715400183-A
Country	US
Kind code	B2
Filing date	Jan 6, 2017
Priority date	Oct 25, 2013
Publication date	Dec 5, 2017
Grant date	Dec 5, 2017

Abstract

A plurality of spectra reflected from one or more substrates at a plurality of different positions on the one or more substrates are represented in the form of a first matrix, and the first matrix is decomposed into products of at least two component matrixes of a first set of component matrixes. The dimensions of each of the at least two component matrixes is reduced to produce a second set of component matrixes containing the at least two matrixes with reduced dimensions.

First claim

What is claimed is:

1. A method of polishing comprising:
polishing a substrate with a rotating polishing pad in a polishing system;
measuring a multiplicity of spectra of light reflected from the substrate with an in-situ spectrographic monitoring system, the multiplicity of spectra including a plurality of spectra collected during a single rotation of the polishing pad from at a plurality of different positions on the substrate, the plurality of spectra represented in the form of a first matrix, each spectrum of the plurality of spectra being represented by a dataset arranged in a single row or a single column of the first matrix;
decomposing the first matrix into products of at least two component matrixes of a first set of component matrixes;
reducing dimensions of each of the at least two component matrixes to produce a second set of component matrixes containing the at least two matrixes with reduced dimensions;
generating a second matrix by taking a product of the matrixes of the second set of component matrixes, the second matrix having the same dimensions as the first matrix, and each single row or each single column of the second matrix comprising a modified dataset representing a modified spectrum corresponding to the spectrum represented by a respective single row or single column of the first matrix; and
controlling the polishing system based on the second matrix.

2. The method of claim 1 , comprising generating a characterizing value based on the second matrix, the characterizing value being associated with a property of the substrate, and controlling the polishing operation based on the characterizing value.

3. The method of claim 1 , comprising grouping spectra from the multiplicity of spectra into different groups, each different group of spectra including a plurality of spectra collected during a single rotation of the polishing pad and represented by a first matrix.

4. The method of claim 3 , wherein the plurality of spectra within a group are reflected from regions of an outer layer of the substrate that have substantially the same thickness, and the different groups have spectra reflected from regions of the outer layer of the substrate that have different thicknesses.

5. The method of claim 1 , wherein decomposing the first matrix comprises applying singular value decomposition, applying CUR matrix approximation, or applying principal component analysis.

6. A polishing system, comprising:
a rotatable support to hold a polishing pad;
a carrier head to hold a substrate in contact with the polishing pad;
a motor to rotate the support;
an in-situ spectrographic monitoring system configured to measure a multiplicity of spectra of light reflected from the substrate during polishing; and
a controller configured to
receive the multiplicity of spectra of light from the in-situ spectrographic monitoring system,
collect, from the multiplicity of spectra, a plurality of spectra collected during a single rotation of the polishing pad from at a plurality of different positions on the substrate, the plurality of spectra represented in the form of a first matrix, each spectrum of the plurality of spectra being represented by a dataset arranged in a single row or a single column of the first matrix,
decompose the first matrix into products of at least two component matrixes of a first set of component matrixes,
reduce dimensions of each of the at least two component matrixes to produce a second set of component matrixes containing the at least two matrixes with reduced dimensions,
generate a second matrix by taking a product of the matrixes of the second set of component matrixes, the second matrix having the same dimensions as the first matrix, and each single row or each single column of the second matrix comprising a modified dataset representing a modified spectrum corresponding to the spectrum represented by a respective single row or single column of the first matrix, and
control the polishing system based on the second matrix.

7. The system of claim 6 , wherein the controller is configured to generate a characterizing value based on the second matrix, the characterizing value being associated with a property of the substrate, and to control the polishing system based on the characterizing value.

8. The system of claim 7 , wherein the controller is configured to determine a polishing endpoint for the substrate based on the characterizing value.

9. The system of claim 6 , wherein the controller is configured to group spectra from the multiplicity of spectra into different groups, each different group of spectra including a plurality of spectra collected during a single rotation of the polishing pad and represented by a first matrix.

10. The system of claim 9 , wherein the plurality of spectra within a group are reflected from regions of an outer layer of the substrate that have substantially the same thickness, and the different groups have spectra reflected from regions of the outer layer of the substrate that have different thicknesses.

11. The system of claim 6 , wherein the controller is configured to decompose the first matrix by applying singular value decomposition, CUR matrix approximation, or principal component analysis.

12. The system of claim 6 , wherein the controller is configured to reduce dimensions of each of the at least two component matrixes by to i) truncate one or more columns and/or one or more rows of each of the at least two component matrixes, or ii) replace all non-zero values of one or more columns or rows of matrix elements with zeros.

13. The system of claim 12 , wherein the controller is configured to truncate or replace non-zero values of one or more columns or rows in which nonzero matrix elements have a value smaller than a predetermined percentage of values of all nonzero matrix elements of the diagonal matrix.

14. The system of claim 13 , wherein the predetermined percentage is 80% or higher.

15. The system of claim 12 , wherein the controller is configured to truncate or replace non-zero values of columns or rows by eliminating or retaining a predetermined number of columns or rows.

Assignees

• Applied Materials Inc

Inventors

• David Jeffrey Drue
• Swedek Boguslaw A
• Cherian Benjamin

Classifications

• H10P74/238

  comprising acting in response to an ongoing measurement without interruption of processing, e.g. endpoint detection or in-situ thickness measurement · CPC title

• H10P74/203

  Structural properties, e.g. testing or measuring thicknesses, line widths, warpage, bond strengths or physical defects · CPC title

• H10P74/23

  characterised by multiple measurements, corrections, marking or sorting processes · CPC title

• H10P52/402 Primary

  of semiconductor materials · CPC title

• H01L21/30625

  Electricity · mapped topic