# US 2026/0276559 A1 · Inspection Method, Inspection Device, and Substrate Processing Device

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

- Number: US 2026/0276559 A1
- Document ID: US20260276559A1
- Status: Published application
- Kind: A1
- Application number: 19/676,631
- Filed: 2026-05-13
- Published: 2026-09-17
- Earliest claimed: 2023-11-24
- Assignees: Tokyo Electron Limited
- Applicants: Tokyo Electron Limited
- Inventors: Hirokazu UEDA; Yasutoshi UMEHARA; Hiroyuki KOISHI
- Web page: https://patentgrep.com/patent/US20260276559A1

## CPC

- G01N 21/8422: {Investigating thin films, e.g. matrix isolation method}
- G01N 21/9501: {Semiconductor wafers (manufacturing processes per se of semiconductor devices implementing a measuring step H10P74/20)}
- G01N 21/211: {Ellipsometry (optical thickness measurement G01B11/06)}
- G01N 21/55: Specular reflectivity
- G01N 21/8806: {Specially adapted optical and illumination features}
- G01N 21/8851: {Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges (G01N21/8806 and G01N21/93 - G01N21/95692 take precedence; optical measurement of dimensions G01B11/00; optical scanning G02B26/10; image transformation G06T3/00; computerised image enhancement G06T5/00; image processing per se for flaw detection G06T7/0002)}
- G01N 21/9505: {Wafer internal defects, e.g. microcracks}
- G01N 21/95607: {using a comparative method}
- G01N 21/95684: {Patterns showing highly reflecting parts, e.g. metallic elements}
- H10P 74/00: Testing or measuring during manufacture or treatment of wafers, substrates or devices
- G01N 2021/213: {Spectrometric ellipsometry}
- G01N 2021/3568: {applied to semiconductors, e.g. Silicon}
- G01N 2021/8427: {Coatings}
- G01N 2021/8438: {Mutilayers}
- G01N 2021/8883: {involving the calculation of gauges, generating models}
- G01N 2021/95615: {with stored comparision signal}
- G01N 2201/1293: resolving multicomponent spectra
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- H10P: GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10

## Abstract

A method for inspecting a substrate comprises measuring a characteristic spectrum indicating light absorption or reflection with respect to wavelength of light by irradiating a measurement area on the substrate with light. An arrangement state of the substrate in the measurement area is determined from a structure of the substrate identified by comparing an inspection target spectrum, which is the measured characteristic spectrum, with a reference spectrum. The reference spectrum is obtained from a training model based on a film thickness or a substrate thickness identified from the substrate. The training model is generated by learning characteristic spectra for a plurality of substrates in association with the film thickness of a film formed on a surface of the substrate or the substrate thickness, in a case where the characteristic spectrum changes according to the structure of the substrate in the measurement area.

## Claims

### Claim 1

A method for inspecting a substrate by irradiating a substrate with light, comprising:
irradiating a measurement area set on the substrate with light and measuring a characteristic spectrum indicating characteristics of light absorption or reflection with respect to a wavelength of the light; and
determining an arrangement state of the substrate in the measurement area from a structure of the substrate identified by comparing an inspection target spectrum, which is the characteristic spectrum measured in said measuring the characteristic spectrum, with a reference spectrum, which is the characteristic spectrum obtained from a training model based on a film thickness or a substrate thickness identified from the substrate from which the inspection target spectrum was acquired, using the training model generated by learning the characteristic spectrum for multiple substrates in association with the film thickness when a film is formed on a surface of the substrate or a substrate thickness in a case where the characteristic spectrum changes according to the structure of the substrate in the measurement area.

### Claim 2 (depends on 1)

The method of claim 1, wherein the characteristic spectrum is any one of an intensity spectrum, an absorption intensity spectrum, an absorbance spectrum, and a reflectance spectrum of reflected light from the measurement area.

### Claim 3 (depends on 1)

The method of claim 1, wherein the light irradiated onto the measurement area is measurement light for performing spectroscopic measurement of a film on the surface of the substrate on which an uneven pattern, which is the structure of the substrate, is formed, and
in said determining the arrangement state of the substrate, whether the measurement area is located within an area where the uneven pattern is formed is determined from a comparison result between the inspection target spectrum obtained by identifying the film thickness and the reference spectrum, using the training model generated by learning the characteristic spectrum formed according to the uneven pattern in association with the film thickness of the film,
when it is determined in said determining the arrangement state that the measurement area is not located within the area where the uneven pattern is formed, moving the measurement area, said measuring the characteristic spectrum and said determining the arrangement state are repeated.

### Claim 4 (depends on 3)

The method of claim 3, wherein the uneven pattern is formed in an inspection area where the spectroscopic measurement is performed.

### Claim 5 (depends on 4)

The method of claim 4, wherein an adjacent area, where an uneven pattern different from the uneven pattern of the inspection area is formed, is provided adjacent to the inspection area of the substrate,
in said determining the arrangement state of the substrate, whether the measurement area is located within the inspection area or in the adjacent area is determined from a comparison result between the inspection target spectrum obtained by identifying the film thickness and the reference spectrum, using the training model generated by learning the characteristic spectrum formed according to the uneven patterns of the inspection area and the adjacent area, and
when it is determined in said determining the arrangement state that the measurement area is located within the adjacent area, in said moving the measurement area, the measurement area is moved based on a direction and a movement distance preset based on an arrangement relationship between the inspection area and the adjacent area.

### Claim 6 (depends on 3)

The method of claim 3, wherein the uneven pattern is an alignment mark provided to perform alignment adjustment of the substrate and formed in a preset area in the surface of the substrate.

### Claim 7 (depends on 3)

The method of claim 3, wherein, a singular point is defined as a wavelength of the light at a position where multiple types of the characteristic spectra obtained by changing the structure of the substrate intersect, or a wavelength of the light at a position where a difference between values of the multiple types of the characteristic spectra becomes minimum when the multiple types of the characteristic spectra do not intersect,
the method further comprising identifying values at the singular points of the inspection target spectrum and the reference spectrum after said measuring the characteristic spectrum, and
said determining the arrangement state of the substrate is not executed when a difference between the value identified in the inspection target spectrum and the value identified in the reference spectrum is greater than or equal to a preset tolerance.

### Claim 8 (depends on 7)

The method of claim 7, wherein the singular point changes according to the film thickness of the film on the surface of the substrate, and
in said identifying values at the singular points, the values at the singular points corresponding to the film thickness are identified.

### Claim 9 (depends on 1)

The method of claim 1, wherein the substrate is a bonded substrate formed by bonding a first substrate and a second substrate, and the light irradiated onto the measurement area is infrared light for performing measurement of an internal structure of the bonded substrate, and
in said determining the arrangement state of the substrate, whether the first substrate and the second substrate are bonded at an accurate position is determined based on a comparison result between the inspection target spectrum obtained by identifying the thickness of the substrate and the reference spectrum, using the training model generated by learning the characteristic spectrum formed according to a bonding state between the first substrate and the second substrate in association with the thickness of the bonded substrate.

### Claim 10 (depends on 9)

The method of claim 9, wherein the bonding state is misalignment of metal pads that are respectively provided on the first substrate and the second substrate to connect the first substrate and the second substrate to each other.

### Claim 11 (depends on 1)

The method of claim 1, wherein in said measuring the characteristic spectrum, the characteristic spectrum is measured by a hyperspectral camera.

### Claim 12 (depends on 11)

The method of claim 11, wherein a measurement area of the characteristic spectrum by the hyperspectral camera is configured in an elongated rectangular shape having a longitudinal dimension greater than a diameter of the substrate, and
in the measurement using the hyperspectral camera, the measurement area scans an entire surface of the substrate by relatively moving the measurement area in a fixed direction with respect to multiple substrates placed on a stage.

### Claim 13

An apparatus for inspecting a substrate, comprising:
a stage;
an inspection part configured to irradiate a measurement area set on the substrate placed on the stage with light and receive reflected light; and
a control part;
wherein the control part outputs a control signal for executing:
measuring a characteristic spectrum indicating characteristics of light absorption or reflection with respect to a wavelength of the light, by irradiating a measurement area set on the substrate with light; and
determining an arrangement state of the substrate in the measurement area from a structure of the substrate identified by comparing an inspection target spectrum, which is the characteristic spectrum measured in said measuring the characteristic spectrum, with a reference spectrum, which is the characteristic spectrum obtained from a training model based on a film thickness or a substrate thickness identified from the substrate from which the inspection target spectrum was acquired, using the training model generated by learning the characteristic spectrum for multiple substrates in association with the film thickness when a film is formed on a surface of the substrate or a substrate thickness in a case where the characteristic spectrum changes according to the structure of the substrate in the measurement area.

### Claim 14 (depends on 13)

The apparatus of claim 13, wherein the characteristic spectrum is any one of an intensity spectrum, an absorption intensity spectrum, an absorbance spectrum, and a reflectance spectrum of reflected light from the measurement area.

### Claim 15 (depends on 13)

The apparatus of claim 13, wherein the inspection part is configured to move the measurement area, and the light irradiated onto the measurement area is measurement light for performing spectroscopic measurement of a film on the surface of the substrate on which an uneven pattern, which is the structure of the substrate, is formed, and
in said determining the arrangement state of the substrate, the control part determines whether the measurement area is located within an area where the uneven pattern is formed from a comparison result between the inspection target spectrum obtained by identifying the film thickness and the reference spectrum, using the training model generated by learning the characteristic spectrum formed according to the uneven pattern in association with the film thickness of the film,
when it is determined in said determining the arrangement state that the measurement area is not located within the area where the uneven pattern is formed, the control part outputs a control signal for repeatedly executing moving the measurement area, said measuring the characteristic spectrum and said determining the arrangement state.

### Claim 16 (depends on 15)

The apparatus of claim 15, wherein the uneven pattern is formed in an inspection area where the spectroscopic measurement is performed.

### Claim 17 (depends on 16)

The apparatus of claim 16, wherein an adjacent area, where an uneven pattern different from the uneven pattern of the inspection area is formed, is provided adjacent to the inspection area of the substrate, and
in said determining the arrangement state of the substrate, the control part determines whether the measurement area is located within the inspection area or in the adjacent area from a comparison result between the inspection target spectrum obtained by identifying the film thickness and the reference spectrum, using the training model generated by learning the characteristic spectrum formed according to the uneven pattern of the inspection area and the adjacent area, and
when it is determined in said determining the arrangement state that the measurement area is located within the adjacent area, the control part outputs a control signal for moving the measurement area based on a direction and a movement distance preset based on an arrangement relationship between the inspection area and the adjacent area in said moving the measurement area.

### Claim 18 (depends on 15)

The apparatus of claim 15, wherein the uneven pattern is an alignment mark provided to perform alignment adjustment of the substrate and formed in a preset area in the surface of the substrate.

### Claim 19 (depends on 15)

The apparatus of claim 15, wherein, a singular point is defined as a wavelength of the light at a position where multiple types of the characteristic spectra obtained by changing the structure of the substrate intersect, or a wavelength of the light at a position where a difference between values of the multiple types of the characteristic spectra becomes minimum when the multiple types of the characteristic spectra do not intersect,
the control part outputs a control signal to identify values at the singular points of the inspection target spectrum and the reference spectrum after said measuring the characteristic spectrum, and to prevent execution of said determining the arrangement state of the substrate when a difference between the value identified in the inspection target spectrum and the value identified in the reference spectrum is greater than or equal to a preset tolerance.

### Claim 20 (depends on 19)

The apparatus of claim 19, wherein the singular point changes according to the film thickness of the film on the surface of the substrate, and
in said identifying values at the singular points, the control part outputs a control signal for identifying the values at the singular points corresponding to the film thickness.

### Claim 21 (depends on 13)

The apparatus of claim 13, wherein the substrate is a bonded substrate formed by bonding a first substrate and a second substrate, and the light irradiated onto the measurement area is infrared light for performing measurement of an internal structure of the bonded substrate, and
in said determining the arrangement state of the substrate, the control part outputs a control signal for determining whether the first substrate and the second substrate are bonded at an accurate position, based on a comparison result between the inspection target spectrum obtained by identifying the thickness of the substrate and the reference spectrum, using the training model generated by learning the characteristic spectrum formed according to a bonding state between the first substrate and the second substrate in association with the thickness of the bonded substrate.

### Claim 22 (depends on 21)

The apparatus of claim 21, wherein the bonding state is misalignment of metal pads that are respectively provided on the first substrate and the second substrate to connect the first substrate and the second substrate to each other.

### Claim 23 (depends on 13)

The apparatus of claim 13, wherein the inspection part includes a hyperspectral camera configured to measure the inspection target spectrum.

### Claim 24 (depends on 23)

The apparatus of claim 23, wherein a measurement area of the hyperspectral camera is configured in an elongated rectangular shape having a longitudinal dimension greater than the diameter of the substrate, and
the measurement area scans an entire surface of the substrate by relatively moving the measurement area in a fixed direction with respect to multiple substrates placed on the stage.

### Claim 25 (depends on 24)

A substrate processing apparatus comprising:
the apparatus described in claim 24;
a rotation axis that rotates the stage;
a processing chamber accommodating the stage on which the multiple substrates are placed and revolve around the rotation axis; and
a processing gas supply part configured to supply a processing gas to the substrate,
wherein the substrate is inspected after processing is performed by supplying the processing gas to the substrate placed on the stage.
