# US 12,020,159 B2 · Training spectrum generation for machine learning system for spectrographic monitoring

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,020,159 B2
- Document ID: US12020159B2
- Status: Granted
- Kind: B2
- Application number: 18/310,520
- Filed: 2023-05-01
- Published: 2023-08-24
- Granted: 2024-06-25
- Earliest claimed: 2018-06-28
- Assignees: Applied Materials, Inc.
- Applicants: Applied Materials, Inc.
- Inventors: Benjamin Cherian; Nicholas A. Wiswell; Jun Qian; Thomas H. Osterheld
- Web page: https://patentgrep.com/patent/US12020159B2

## CPC

- G06N 3/08: Learning methods
- G05B 13/027: {using neural networks only}
- G05B 19/4063: Monitoring general control system (G05B19/4062 takes precedence)
- G05B 19/4155: characterised by program execution, i.e. part program or machine function execution, e.g. selection of a program
- G06N 3/045: Combinations of networks
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/0475: Generative networks
- G06N 3/09: Supervised learning
- G06N 3/094: Adversarial learning
- H10P 74/203: {Structural properties, e.g. testing or measuring thicknesses, line widths, warpage, bond strengths or physical defects}
- G05B 2219/32335: Use of ann, neural network
- G05B 2219/40066: Stack and align identical layers, laminates, electronic substrate layers
- G05B 2219/41054: Using neural network techniques
- G05B 2219/45031: Manufacturing semiconductor wafers
- G05B 2219/45199: Polish
- G06N 3/047: Probabilistic or stochastic networks
- G06N 3/048: Activation functions
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- H10P: GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10

## Abstract

A method of generating training spectra for training of a neural network includes measuring a first plurality of training spectra from one or more sample substrates, measuring a characterizing value for each training spectra of the plurality of training spectra to generate a plurality of characterizing values with each training spectrum having an associated characterizing value, measuring a plurality of dummy spectra during processing of one or more dummy substrates, and generating a second plurality of training spectra by combining the first plurality of training spectra and the plurality of dummy spectra, there being a greater number of spectra in the second plurality of training spectra than in the first plurality of training spectra. Each training spectrum of the second plurality of training spectra having an associated characterizing value.

## Claims

### Claim 1

A method of controlling processing of a substrate, comprising:
measuring a first plurality of training spectra from one or more sample substrates;
measuring a characterizing value for each training spectrum of the first plurality of training spectra to generate a plurality of characterizing values with each training spectrum having an associated characterizing value;
measuring a plurality of dummy spectra during processing of one or more dummy substrates;
generating a second plurality of training spectra by combining the first plurality of training spectra and the plurality of dummy spectra, there being a greater number of spectra in the second plurality of training spectra than in the first plurality of training spectra, each training spectrum of the second plurality of training spectra having an associated characterizing value; and
training an artificial neural network using the second plurality of training spectra, the artificial neural network having a plurality of input nodes to a plurality of spectral values, an output node to output the characterizing value, and a plurality of hidden nodes connecting the input nodes to the output node;
receiving, from an in-situ optical monitoring system, a measured spectrum of light reflected from a substrate undergoing processing that modifies a thickness of an outer layer of the substrate;
generating a measured characterizing value for the measured spectrum by applying spectral values from the measured spectrum into the artificial neural network; and
at least one of halting processing of the substrate or adjusting a processing parameter based on the characterizing value.

### Claim 2 (depends on 1)

The method of claim 1, wherein the processing comprises chemical mechanical polishing.

### Claim 3 (depends on 1)

The method of claim 1, wherein the dummy substrates comprises blank semiconductor substrates.

### Claim 4 (depends on 1)

The method of claim 1, comprising normalizing the plurality of dummy spectra to generate a plurality of normalized dummy spectra.

### Claim 5 (depends on 4)

The method of claim 4, wherein combining the first plurality of training spectra and the plurality dummy spectra comprises, for each normalized dummy spectrum from the plurality of normalized dummy spectra, multiplying the normalized dummy spectrum by one of the first plurality of training spectra to generate one of the second plurality of training spectra.

### Claim 6 (depends on 5)

The method of claim 5, comprising multiplying the normalized dummy spectrum by a randomly selected one of the first plurality of training spectra.

### Claim 7 (depends on 1)

The method of claim 1, wherein the characterizing value comprises a thickness of an outermost layer of the substrate.

### Claim 8

A computer program product for controlling processing of a substrate, the computer program product tangibly embodied in a non-transitory computer readable media and comprising instructions for causing a processor to:
receive a first plurality of training spectra and plurality of characterizing values, each training spectrum of the first plurality of training spectra having an associated characterizing value from the plurality of characterizing values;
receive a plurality of dummy spectra representing spectra obtained from processing of one or more dummy substrates;
generate a second plurality of training spectra by combining the first plurality of training spectra and the plurality of dummy spectra, there being a greater number of spectra in the second plurality of training spectra than in the first plurality of training spectra, each training spectrum of the second plurality of training spectra having an associated characterizing value;
train an artificial neural network using the second plurality of training spectra, the artificial neural network having a plurality of input nodes to a plurality of spectral values, an output node to output the characterizing value, and a plurality of hidden nodes connecting the input nodes to the output node;
receive, from an in-situ optical monitoring system, a measured spectrum of light reflected from a substrate undergoing processing that modifies a thickness of an outer layer of the substrate;
generate a measured characterizing value for the measured spectrum by applying spectral values from the measured spectrum into the artificial neural network; and
at least one of halt processing of the substrate or adjust a processing parameter based on the characterizing value.

### Claim 9 (depends on 8)

The computer program product of claim 8, comprising instructions to reduce a dimensionality of the second plurality of training spectra to generate a plurality of spectral training values and to apply the spectra training values and the characterizing values associated with the second plurality of training spectra to the artificial neural network to train the artificial neural network.

### Claim 10 (depends on 8)

The computer program product of claim 8, wherein the instructions to generate the measured characterizing value comprise instructions to reduce a dimensionality of the measured spectrum to generate the spectral values.

### Claim 11 (depends on 8)

The computer program product of claim 8, wherein the measured characterizing value comprises a thickness of an outermost layer of the substrate.

### Claim 12 (depends on 8)

The computer program product of claim 8, wherein the processing comprises chemical mechanical polishing.