# US 2022/0035236 A1 · Method of forming shape on mask based on deep learning, and mask manufacturing method using the method of forming the shape on mask

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

- Number: US 2022/0035236 A1
- Document ID: US20220035236A1
- Status: Published application
- Kind: A1
- Application number: 17/154,279
- Filed: 2021-01-21
- Published: 2022-02-03
- Earliest claimed: 2020-07-29
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Useong KIM; Mincheol KANG; Woojoo SIM
- Web page: https://patentgrep.com/patent/US20220035236A1

## CPC

- G03F 1/36: Masks having proximity correction features; Preparation thereof, e.g. optical proximity correction [OPC] design processes
- G03F 1/70: Adapting basic layout or design of masks to lithographic process requirements, e.g., second iteration correction of mask patterns for imaging
- G03F 7/70308: {Optical correction elements, filters or phase plates for manipulating imaging light, e.g. intensity, wavelength, polarisation, phase or image shift}
- G03F 7/70441: {Optical proximity correction [OPC]}
- G06N 3/045: Combinations of networks
- G06N 3/047: Probabilistic or stochastic networks
- G06N 3/0475: Generative networks
- G06N 3/08: Learning methods
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- H10P 76/00: Manufacture or treatment of masks on semiconductor bodies, e.g. by lithography or photolithography
- 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
- H10P: GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10

## Abstract

Provided are a method of forming a mask, the method accurately and quickly restoring an image on the mask to the shape on the mask, and a mask manufacturing method using the method of forming the mask. The method of forming a mask includes obtaining first images by performing rasterization and image correction on shapes on the mask corresponding to first patterns on a wafer, obtaining second images by applying a transformation to the shapes on the mask, performing deep learning based on a transformation relationship between ones of the first images and ones of the second images corresponding to the first images, and forming a target shape on the mask corresponding to a target pattern on the wafer, based on the deep learning. The mask is manufactured based on the target shape on the mask.

## Claims

### Claim 1

A method of forming a mask, the method comprising:
obtaining first images by performing rasterization and image correction on shapes on the mask corresponding to first patterns on a wafer;
obtaining second images by applying a transformation to the shapes on the mask;
performing deep learning based on a transformation relationship between ones of the first images and ones of the second images corresponding to the first images;
forming a target shape on the mask corresponding to a target pattern on the wafer, based on the deep learning; and
manufacturing the mask based on the target shape on the mask.

### Claim 2 (depends on 1)

The method of claim 1, wherein the forming of the target shape on the mask comprises:
transforming a shape of the shapes on the mask corresponding to the target pattern on the wafer into a target first image through rasterization and image correction;
transforming the target first image into a target second image, based on the deep learning; and
transforming the target second image into the target shape on the mask.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
before the transforming the shape of the shapes into the target first image, designing an initial shape on the mask corresponding to the target pattern,
wherein the transforming the shape into the target first image comprises:
generating an initial first image through low-pass filtering, down sampling, and parametrization based on the initial shape; and
generating the target first image by performing image correction on the initial first image.

### Claim 4 (depends on 2)

The method of claim 2,
wherein the rasterization comprises low-pass filtering, down sampling, and parametrization, and
wherein after the transforming into the target shape, additional image correction is not performed.

### Claim 5 (depends on 2)

The method of claim 2,
wherein the target first image has a bitmap format, and
wherein the target second image has a bitmap format to which interpolation is applied.

### Claim 6 (depends on 5)

The method of claim 5,
wherein the transformation comprises a signed distance computation, and
wherein the target second image comprises a signed distance image.

### Claim 7 (depends on 1)

The method of claim 1,
wherein the first images and the second images have a resolution of several nanometers (nm) or greater, and
wherein the target shape has a resolution of about 0.1 nm.

### Claim 8 (depends on 1)

The method of claim 1, wherein the target shape has a vector format and includes a curvilinear shape.

### Claim 9 (depends on 1)

The method of claim 1, wherein the deep learning uses a generative neural network algorithm.

### Claim 10 (depends on 1)

The method of claim 1, wherein the target shape is transmitted as Mask Tape-Out (MTO) design data and used in the manufacturing of the mask.

### Claim 11

A method of forming a mask, the method comprising:
designing initial shapes on the mask corresponding to first patterns on a wafer;
obtaining first images that have a bitmap format by performing rasterization and image correction on the initial shapes;
obtaining signed distance images by applying a signed distance computation to the initial shapes on the mask corresponding to the first patterns on the wafer;
performing deep learning with respect to a transformation relationship between ones of the first images and ones of the signed distance images corresponding to the first images;
transforming an initial shape of the initial shapes on the mask corresponding to a target pattern on the wafer into a target first image by performing rasterization and image correction on the initial shape on the mask corresponding to the target pattern on the wafer;
transforming the target first image into a target signed distance image, based on the deep learning;
transforming the target signed distance image into a target shape on the mask; and
manufacturing the mask comprising the target shape on the mask.

### Claim 12 (depends on 11)

The method of claim 11,
wherein the rasterization comprises low-pass filtering, down sampling, and parametrization, and
wherein after the transforming the target signed distance image into the target shape, additional image correction is not performed.

### Claim 13 (depends on 11)

The method of claim 11,
wherein the target signed distance image has a bitmap format to which interpolation is applied, and
wherein the target shape has a vector format and includes a curvilinear shape.

### Claim 14 (depends on 11)

The method of claim 11, wherein the deep learning uses a generative neural network algorithm.

### Claim 15

A mask manufacturing method comprising:
obtaining first images by performing a first rasterization and a first image correction on shapes on a mask corresponding to first patterns on a wafer;
obtaining second images by applying a transformation to the shapes on the mask;
performing deep learning with respect to a transformation relationship between ones of the first images and ones of the second images corresponding to the first images;
forming a target shape on the mask corresponding to a target pattern on the wafer, based on the deep learning;
transmitting the target shape as Mask Tape-Out (MTO) design data;
preparing for mask data, based on the MTO design data; and
performing exposure on a mask substrate of the mask, based on the mask data.

### Claim 16 (depends on 15)

The mask manufacturing method of claim 15, wherein the forming of the target shape on the mask comprises:
transforming a shape of the shapes on the mask corresponding to the target pattern on the wafer into a target first image through a second rasterization and a second image correction;
transforming the target first image into a target second image, based on the deep learning; and
transforming the target second image into the target shape on the mask.

### Claim 17 (depends on 16)

The mask manufacturing method of claim 16,
wherein the second rasterization comprises low-pass filtering, down sampling, and parametrization, and
wherein after the transforming the target second image into the target shape, additional image correction is not performed.

### Claim 18 (depends on 16)

The mask manufacturing method of claim 16,
wherein the target first image has a bitmap format,
wherein the target second image has a bitmap format to which interpolation is applied,
wherein the transformation to the shapes is a signed distance computation, and
wherein the target second image is a signed distance image.

### Claim 19 (depends on 15)

The mask manufacturing method of claim 15, wherein the target shape has a vector format and includes a curvilinear shape.

### Claim 20 (depends on 15)

The mask manufacturing method of claim 15, wherein the deep learning uses a generative neural network algorithm.
