Metrology Source Separation for True Value Identification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In lithographic processes, determining the accuracy and precision of substrate measurement recipes is challenging due to the combination of true values and systematic errors in measurement results, making it difficult to isolate the contribution of each source.

Innovation Solution

A method is developed to determine metrology contributions from statistically independent sources by providing multiple contributions at various measurement settings, using statistical methods like Independent Component Analysis (ICA) to identify the contribution with least dependence on measurement settings, which is assumed to be the true value.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple measurement recipes are used to improve measurement accuracy, then the ability to identify true values and systematic errors is improved, but the complexity of the measurement process increases

Engineering Contradiction:

Improvemeasurement accuracyVSAvoidmeasurement process complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The measurement results are segmented into contributions from different statistically independent sources using Independent Component_analysis. This separates the true value from systematic errors by analyzing how each source contributes differently across multiple measurement recipes, allowing identification of the true value as the contribution with least dependence on measurement settings.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The problem is transformed from a one-dimensional measurement value to a multi-dimensional analysis by considering multiple measurement recipes with different settings. By adding the dimension of measurement settings and analyzing contributions across this dimension, the method identifies the true value as the component that remains consistent regardless of the measurement settings used.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Measurement precision

If multiple measurement settings are used to separate contributions from different sources, then the identification of true values is improved, but the measurement time increases

Engineering Contradiction:

Improvetrue value identificationVSAvoidmeasurement time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The method performs preliminary analysis by collecting measurement results from multiple recipes with different settings before attempting to identify the true value. This preliminary data collection across various settings enables the subsequent separation of contributions using statistical methods, making the identification process more efficient than trial-and-error approaches.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of physically repeating measurements multiple times under identical conditions, the method uses copies of measurement recipes with different settings. By analyzing how the same physical quantity is measured under different conditions, the true value can be identified as the consistent contribution across all these 'copies' of measurements.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20250004385A1Source separation from metrology data

Publication Date: 2025.01.02 ASML NETHERLANDS BV

AI summary

A method to determine a metrology contribution from statistically independent sources, the method including providing a plurality of contributions from statistically independent sources obtained at a plurality of measurement settings, and determining a metrology contribution from the contributions wherein the metrology contribution is the contribution having least dependence as a function of the measurement settings.