EUV Mask Inspection with Deep-Learned Design-Based References

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing EUV mask inspection devices face challenges in maintaining high inspection accuracy due to instrumental errors between different inspection devices and changes in the mask itself, particularly when using extreme ultraviolet light, which is difficult to correct with conventional golden sharing methods.

Innovation Solution

An EUV mask inspection device and method utilizing a learning machine for deep learning to generate a conversion model based on captured images and design data-derived gray images, allowing for the generation of reference images and accurate comparisons to improve inspection accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional golden sharing methods are used for EUV mask inspection, then inspection can be performed across multiple devices, but inspection accuracy deteriorates due to instrumental errors between devices and mask changes

Engineering Contradiction:

Improveinspection device compatibilityVSAvoidinspection accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent creates a virtual copy of the mask pattern by generating a reference image from design data through optical simulation. This reference image serves as a perfect template that can be compared with captured images from any inspection device, eliminating the need for physical master masks and enabling accurate inspections across multiple devices without instrumental error accumulation

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent transforms the inspection approach by changing from direct physical comparison (golden sharing) to a parameter-based comparison using simulated optical characteristics. The reference image is generated by simulating optical parameters (aberration, focus, illumination) to match the actual inspection conditions, allowing accurate defect detection regardless of the specific inspection device used

Inventive Principle:

Principle #35Parameter changes

2Difficulty of detecting and measuring

If high-resolution EUV light is used for mask inspection, then detection capability improves, but instrumental errors and mask changes become more difficult to correct

Engineering Contradiction:

Improvedefect detection capabilityVSAvoidinspection accuracy

Core Design Contradiction:

Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The patent performs preliminary action by pre-calculating and pre-simulating the optical characteristics and mask pattern before actual inspection. The reference image is generated in advance using design data and optical simulation that mirrors the actual inspection conditions, so that when high-resolution EUV images are captured, the comparison is already optimized for maximum accuracy without needing complex real-time corrections

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If design data is used to generate reference images, then inspection accuracy improves by eliminating instrumental errors, but system complexity and cost increase

Engineering Contradiction:

Improveinspection accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical physical mask comparison system with a digital simulation and image processing system. Instead of requiring physical master masks and complex alignment mechanisms, the system uses computational optical simulation to generate reference images from design data, then compares these with captured images using digital image processing algorithms, significantly simplifying the overall system architecture

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12361536B2EUV mask inspection device, EUV mask inspection method, non-transitory computer-readable medium storing EUV mask inspection program, and EUV mask inspection system including a gray image based on design data of a pattern

Publication Date: 2025.07.15 LASERTEC CORP

AI summary

An inspection device according to one aspect of the present disclosure includes an image capturing unit configured to capture an image of an EUV mask provided with a pattern, a storage unit configured to store a database intermediate file including a gray image obtained by pixelating a binarized image rasterized from design data of the pattern, and a processing unit configured to inspect the EUV mask on the basis of a captured image obtained by the image capturing unit capturing an image of the EUV mask. The processing unit includes a conversion model generated by a learning machine configured to perform learning by deep learning, a reference image generation unit configured to generate a reference image from the gray image by using the conversion model, and a comparison unit configured to compare the reference image with the captured image.