# US 2022/0035237 A1 · Process proximity correction method and the computing device for the same

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

- Number: US 2022/0035237 A1
- Document ID: US20220035237A1
- Status: Published application
- Kind: A1
- Application number: 17/245,947
- Filed: 2021-04-30
- Published: 2022-02-03
- Earliest claimed: 2020-07-29
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Soo Yong LEE; Min-Cheol KANG; U Seong KIM; Seung Hune YANG; Jee Yong LEE
- Web page: https://patentgrep.com/patent/US20220035237A1

## CPC

- G03F 1/36: Masks having proximity correction features; Preparation thereof, e.g. optical proximity correction [OPC] design processes
- G03F 1/76: Patterning of masks by imaging
- G06F 30/3323: using formal methods, e.g. equivalence checking or property checking
- G06F 30/392: Floor-planning or layout, e.g. partitioning or placement
- G06T 3/4046: using neural networks
- G06T 7/0006: {using a design-rule based approach}
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30148: Semiconductor; IC; Wafer
- 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)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A process proximity correction method is performed by a process proximity correction computing device which performs a process proximity correction (PPC) through at least one of a plurality of processors. The process proximity correction method includes: converting a target layout including a plurality of patterns into an image, zooming-in or zooming-out the image at a plurality of magnifications to generate a plurality of input channels, receiving the plurality of input channels and performing machine learning to predict an after-cleaning image (ACI), comparing the predicted after-cleaning image with a target value to generate an after-cleaning image error, and adjusting the target layout on the basis of the after-cleaning image error.

## Claims

### Claim 1

A process proximity correction method performed by a process proximity correction computing device which performs a process proximity correction (PPC) through at least one of a plurality of processors, the process proximity correction method comprising:
converting a target layout including a plurality of patterns into an image;
zooming-in or zooming-out the image at a plurality of magnifications to generate a plurality of input channels;
receiving the plurality of input channels and performing machine learning to predict an after-cleaning image (ACI);
comparing the predicted after-cleaning image with a target value to generate an after-cleaning image error; and
adjusting the target layout on the basis of the after-cleaning image error.

### Claim 2 (depends on 1)

The process proximity correction method of claim 1, wherein the machine learning includes deep learning.

### Claim 3 (depends on 2)

The process proximity correction method of claim 2, wherein the deep learning uses a generative adversarial network (GAN).

### Claim 4 (depends on 2)

The process proximity correction method of claim 2, wherein the deep learning uses a convolutional neural network (CNN).

### Claim 5 (depends on 2)

The process proximity correction method of claim 2, wherein the deep learning uses an artificial neural network (ANN).

### Claim 6 (depends on 1)

The process proximity correction method of claim 1, wherein, when the after-cleaning image error becomes 0, adjustment of the target layout is stopped.

### Claim 7 (depends on 1)

The process proximity correction method of claim 1, wherein the machine learning includes receiving of information on the plurality of patterns to predict the after-cleaning image.

### Claim 8 (depends on 7)

The process proximity correction method of claim 7, wherein information on the plurality of patterns includes the number of the plurality of patterns.

### Claim 9 (depends on 7)

The process proximity correction method of claim 7, wherein information on the plurality of patterns includes areas of each of the plurality of patterns.

### Claim 10 (depends on 1)

The process proximity correction method of claim 1, wherein conversion into the image uses a binary type.

### Claim 11 (depends on 1)

The process proximity correction method of claim 1, wherein conversion into the image uses an overlap type.

### Claim 12 (depends on 1)

The process proximity correction method of claim 1, wherein conversion into the image uses a level set type.

### Claim 13 (depends on 1)

The process proximity correction method of claim 1, wherein prediction of the after-cleaning image includes prediction of a critical dimension (CD) of the after-cleaning image.

### Claim 14 (depends on 1)

The process proximity correction method of claim 1, further comprising:
predicting a plurality of after-cleaning images using the plurality of received input channels, respectively; and
based on the plurality of after-cleaning images, determining whether to continue adjust the target layout or to stop further adjustment of the target layout.

### Claim 15

A process proximity correction method of a process proximity correction computing device which performs a process proximity correction (PPC) through at least one of a plurality of processors, the process proximity correction method comprising:
converting a target layout including a plurality of patterns into a first image;
generating a first input channel by obtaining a first zoomed out version of the first image at a first magnification, and generating a second input channel by obtaining a second zoomed out version of the first image at a second magnification;
receiving the first input channel and the second input channel and performing machine learning to predict an after-cleaning image (ACI);
comparing the predicted after-cleaning image with a target value to generate after-cleaning image error; and
adjusting the target layout and converting the target layout into a second image different from the first image, on the basis of the after-cleaning image error.

### Claim 16 (depends on 15)

The process proximity correction method of claim 15, wherein when the after-cleaning image error becomes 0, adjustment of the target layout is stopped.

### Claim 17 (depends on 15)

The process proximity correction method of claim 15, wherein the machine learning includes receiving of information on the plurality of patterns to predict the after-cleaning image.

### Claim 18 (depends on 15)

The process proximity correction method of claim 15, wherein prediction of the after-cleaning includes prediction of a critical dimension (CD) of the after-cleaning image.

### Claim 19 (depends on 15)

The process proximity correction method of claim 15, wherein the machine learning is performed using the first input channel and the second input channel to predict the ACI.

### Claim 20

A process proximity correction computing device which includes a plurality of processors,
wherein at least one of the plurality of processors executes a process proximity correction, and
execution of the process proximity correction includes
converting a target layout including a plurality of patterns into an image,
zooming-in or zooming-out the image at a plurality of magnifications to generate a plurality of input channels,
receiving the plurality of input channels and performing machine learning using the plurality of input channels, to predict after-cleaning image (ACI),
comparing the predicted after-cleaning image with a target value to generate an after-cleaning image error, and
adjusting the target layout on the basis of the after-cleaning image error.

### Claim at position 21

21-27. (canceled)