# US 2021/0334444 A1 · Method and computing device for manufacturing semiconductor device

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

- Number: US 2021/0334444 A1
- Document ID: US20210334444A1
- Status: Published application
- Kind: A1
- Application number: 17/112,048
- Filed: 2020-12-04
- Published: 2021-10-28
- Earliest claimed: 2020-04-28
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Sooyong LEE; Jeeyong LEE; Jaeho JEONG
- Web page: https://patentgrep.com/patent/US20210334444A1

## CPC

- G06F 30/392: Floor-planning or layout, e.g. partitioning or placement
- G03F 1/36: Masks having proximity correction features; Preparation thereof, e.g. optical proximity correction [OPC] design processes
- G03F 7/70441: {Optical proximity correction [OPC]}
- G03F 7/705: {Modelling or simulating from physical phenomena up to complete wafer processes or whole workflow in wafer productions}
- G06N 5/04: Inference or reasoning models
- G06N 20/00: Machine learning
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06F 30/398: Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM] (optical proximity correction [OPC] design processes G03F1/36)
- G06F 2119/18: Manufacturability analysis or optimisation for manufacturability
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G03F: PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR; (phototypographic composing devices B41B; photosensitive materials or processes for photographic purposes G03C; electrophotography, sensitive layers or processes therefor G03G)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Disclosed is a method for fabricating of a semiconductor device. The method includes receiving a first layout including patterns for the fabrication of the semiconductor device, performing machine learning-based process proximity correction (PPC) based on features of the patterns of the first layout to generate a second layout, and performing optical proximity correction (OPC) on the second layout to generate a third layout.

## Claims

### Claim 1

A method for fabricating of a semiconductor device, the method comprising:
receiving a first layout including patterns for the fabrication of the semiconductor device;
performing machine learning-based process proximity correction (PPC) based on features of the patterns of the first layout to generate a second layout; and
performing optical proximity correction (OPC) on the second layout to generate a third layout.

### Claim 2 (depends on 1)

The method of claim 1, wherein the features of the patterns include sizes of the patterns.

### Claim 3 (depends on 1)

The method of claim 1, wherein the features of the patterns include displacement information associated with the patterns.

### Claim 4 (depends on 3)

The method of claim 3, wherein the displacement information includes an influence that neighboring patterns exert on a selected pattern, wherein as a distance between the selected pattern and a neighboring pattern of the patterns increases, the influence that the neighboring pattern exerts on the selected pattern decreases.

### Claim 5 (depends on 4)

The method of claim 4, wherein the decrease in the influence is based on a Gaussian distribution.

### Claim 6 (depends on 1)

The method of claim 1, wherein the features of the patterns include information corresponding to an electric field applied to the patterns during an etching process.

### Claim 7 (depends on 6)

The method of claim 6, wherein the electric field information includes a sum of position vectors of neighboring patterns, the neighboring patterns placed around a first pattern selected from the patterns, the neighboring patterns selected from the patterns.

### Claim 8 (depends on 7)

The method of claim 7, wherein magnitudes of the position vectors decrease as a distance between the selected first pattern and the neighboring patterns increases.

### Claim 9 (depends on 8)

The method of claim 8, wherein the decrease in the magnitudes of the position vectors is based on a Gaussian distribution.

### Claim 10 (depends on 1)

The method of claim 1, wherein the features of the patterns include a skew associated with an etching process.

### Claim 11 (depends on 10)

The method of claim 10, wherein the skew includes harmonics correcting angles of position vectors of neighboring patterns which are placed around a first pattern “m” times (m being a positive integer), the first pattern selected from among the patterns).

### Claim 12 (depends on 1)

The method of claim 1, wherein the generating of the second layout includes:
receiving an image of the first layout;
extracting the features of the patterns from the image of the first layout; and
performing a machine learning-based inference based on the features of the patterns to generate an image corresponding to an after cleaning inspection (ACI).

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
adjusting the features based on a difference between the generated ACI image and the image of the first layout; and
performing the machine learning-based inference based on the adjusted features to generate an adjusted ACI image.

### Claim 14 (depends on 12)

The method of claim 12, wherein the generating of the ACI image includes:
performing a first machine learning-based inference on the features, the first machine learning-based inference based on linear regression; and
performing a second machine learning-based inference on a result of the first machine learning-based inference, the second machine learning-based inference based on nonlinear regression, on a result of the first machine learning-based inference.

### Claim 15

A method for fabricating of a semiconductor device comprising:
receiving an image of a first layout including patterns for the manufacturing of the semiconductor device;
dividing the image of the first layout to generate grids;
extracting features of the grids; and
performing a machine learning-based inference on the features to generate an image associated with an after cleaning inspection (ACI).

### Claim 16 (depends on 15)

The method of claim 15, wherein the features include (a) a size of each of grid patterns included in each of the grids, the grid patterns selected from among the patterns and (b) a density of the grid patterns included in each of the grids.

### Claim 17 (depends on 15)

The method of claim 15, wherein the extracting of the features includes:
extracting an influence as a part of the features, the influence corresponding to an amount by which neighboring grids placed around a grid selected from the grids exert on the selected grid.

### Claim 18 (depends on 17)

The method of claim 17, wherein the influence is extracted by applying an n×n convolution filter (n being a positive integer) to the grids.

### Claim 19 (depends on 15)

The method of claim 15, further comprising:
extracting features of patterns of a first grid, the first grid belonging to a zone of interest from among the grids; and
performing a second machine learning-based inference on the features of the pattern of the grid to generate the image of the ACI.

### Claim 20

A computing device comprising:
processors, wherein at least one processor of the processors is configured to execute machine readable instructions that, when executed by the at least one processor, cause the computing device to perform generation of a layout for manufacturing of a semiconductor device,
wherein the generation of the layout for the manufacturing of the semiconductor device includes,
receiving a first layout including patterns for the manufacturing of the semiconductor device,
performing machine learning-based process proximity correction (PPC) based on features of the patterns of the first layout to generate a second layout, and
performing optical proximity correction (OPC) on the second layout to generate the layout.
