Multi-Die Mask Defect Detection Using Hybrid Reference Images

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current mask inspection methods for EUV masks face challenges in reliably detecting defects due to limitations in die-to-database and die-to-die inspection techniques, particularly with complex optical proximity correction designs and focus control issues, which affect sensitivity and accuracy.

Innovation Solution

A system and method combining aspects of die-to-database and die-to-die inspection by generating a design-based reference image and a die-equivalent reference image, using EUV and DUV wavelengths, along with charged particle beams, to accurately detect defects on multi-die masks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If die-to-database inspection is used, then defect detection can be performed using design reference, but rendering errors from complex OPC designs reduce detection sensitivity

Engineering Contradiction:
Improvedefect detection sensitivityVSAvoidrendering error
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent merges die-to-database inspection with die-to-die inspection by generating a composite reference image that combines design-based reference data with actual measured data from multiple dies. This hybrid approach eliminates rendering errors from complex OPC designs while maintaining the ability to detect systematic writer errors, thereby resolving the contradiction between using design reference and avoiding rendering errors.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces an intermediary step of measuring multiple actual dies and generating a composite reference image that serves as a mediator between the design database and defect detection. This intermediary composite reference eliminates the direct rendering error path while preserving the reference function, allowing accurate defect detection without the harmful rendering errors from complex OPC.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of manufacture

If die-to-die inspection is used, then inspection can be performed without design database, but systematic writer errors cannot be detected and performance is impacted by photomask quality and focus control

Engineering Contradiction:
Improveinspection process simplicityVSAvoiddefect detection reliability
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The patent combines the simplicity of die-to-die inspection with the reliability of design-based reference by measuring multiple actual dies and creating a composite reference image. This merged approach maintains ease of manufacture by using only actual measurements while improving reliability by enabling detection of systematic writer errors through the multi-die comparison methodology.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent makes the inspection system multi-functional by enabling it to detect both random defects and systematic writer errors using the same die-to-die measurement approach. The composite reference image generated from multiple dies serves multiple purposes: it provides a reference for defect detection, compensates for focus variations, and enables systematic error detection, thereby improving reliability without sacrificing process simplicity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Manufacturing precision

If mask patterns are substantially sensitive to focus, then high precision patterning is achieved, but reliable defect detection becomes significantly challenging

Engineering Contradiction:
Improvepatterning precisionVSAvoiddefect detection reliability
Core Design Contradiction:
Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The patent segments the focus variation problem by measuring multiple dies at potentially different focus conditions and then processing these segmented measurements to create a composite reference image. This segmentation allows the system to handle focus variations in sensitive mask patterns while maintaining defect detection reliability through the multi-die comparison approach.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements feedback by using the measured data from multiple dies to refine and update the composite reference image, which is then used for defect detection. This feedback loop compensates for focus variations and aberrations in sensitive mask patterns, maintaining both patterning precision and defect detection reliability by continuously improving the reference based on actual measurements.

Inventive Principle:
Principle #23Feedback

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enhances defect detection sensitivity and reliability by overcoming focus and aberration errors, enabling the detection of systematic writer errors and improving the accuracy of defect identification on EUV masks.

Implementation Method 1

using EUV and DUV wavelengths

Methodology Applied
Scientific EffectElectromagnetic radiation: Light

Implementation Method 2

charged particle beams

Methodology Applied
Scientific EffectCharged particle beam: Electron Beam

Data Source

PatentEP4327082B1Defect detection for multi-die masks
Publication Date: 2025.12.31 KLA CORP
  • EP4327082B1 patent drawingFigure 1
  • EP4327082B1 patent drawingFigure 2
  • EP4327082B1 patent drawingFigure 3~5

AI summary

Methods and systems for detecting defects on a mask are provided. One method includes generating a database reference image for a multi-die mask by simulation and detecting first defects on the mask by comparing the database reference image to images of the mask generated by an imaging subsystem for a first of the multiple dies. The method also includes generating a die reference image for the first of the multiple dies by applying one or more parameters of the imaging subsystem learned by generating the database reference image to the images generated by the imaging subsystem of one or more of the multiple dies other than the first multiple die. In addition, the method includes detecting second defects on the mask by comparing the die reference image to the images of the mask generated by the imaging subsystem for the first of the multiple dies.