Regression-Based Metrology Target Selection for Focus and Dose Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current CD metrology processes for semiconductor fabrication are time-consuming and inefficient, particularly in determining dimension-of-interest (DOI) values and focus sensitivity, often requiring extensive library matching and CD-SEM analysis.

Innovation Solution

A method and system that utilize a focus and exposure matrix wafer with metrology targets, where measurement results are processed through regression to calculate diagnostic parameters like focus sensitivity, library precision, and printability, identifying candidate targets without the need for time-consuming CD-SEM analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If scatterometry/ellipsometry based spectra are collected and stored in a library for CD metrology, then measurement capability is provided, but processing time is excessive due to extensive library matching requirements

Engineering Contradiction:
ImproveCD metrology capabilityVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts and removes the time-consuming library matching step from the CD metrology process. Instead of collecting spectra for all focus-exposure combinations and performing extensive library matching, the invention directly measures selected metrology targets and uses regression analysis to determine DOI values, eliminating the bottleneck of spectral library comparison while maintaining measurement capability

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary selection of metrology targets with specific characteristics (high focus sensitivity, good printability) before the actual measurement process. By pre-identifying targets that will provide the most diagnostic information through regression analysis, the system avoids the need for extensive post-measurement library matching, thus reducing processing time while maintaining measurement precision

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If CD-SEM analysis is performed to determine printability of metrology targets, then accurate printability assessment is achieved, but the process becomes even more time-consuming

Engineering Contradiction:
Improveprintability assessment accuracyVSAvoidanalysis time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the mechanical/physical CD-SEM analysis process with a computational regression-based approach. Instead of using electron microscopy to directly measure and assess printability, the invention uses scatterometry or ellipsometry measurements combined with regression analysis to determine DOI values and calculate printability metrics, achieving similar accuracy without the time-consuming SEM analysis step

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the measurement parameters and methodology from direct physical measurement (CD-SEM) to indirect optical measurement with mathematical analysis (scatterometry/ellipsometry with regression). By measuring optical properties and using regression to extract dimensional information, the system achieves printability assessment without requiring the time-intensive CD-SEM process

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If metrology targets are selected without systematic criteria, then target selection is simple, but diagnostic capability for focus and dose control is insufficient

Engineering Contradiction:
Improvetarget selection simplicityVSAvoiddiagnostic capability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent changes the selection criteria from simple geometric or arbitrary parameters to performance-based parameters that can be evaluated through regression analysis. By calculating focus sensitivity, library precision, and printability metrics for each potential metrology target, the system objectively identifies targets that will provide the best diagnostic capability for focus and dose control, ensuring reliability while maintaining ease of selection through automated calculation

Inventive Principle:
Principle #35Parameter changes

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

This approach provides fast and efficient critical dimension target selection for focus and dose applications by determining DOI values and diagnostic parameters through regression, reducing processing time and improving metrology capabilities.

Implementation Method 1

Scatterometry/ellipsometry based spectra may be collected from test targets

Methodology Applied
Scientific EffectScatterometry: Scattering

Implementation Method 2

Scatterometry/ellipsometry based spectra may be collected from test targets

Methodology Applied
Scientific EffectEllipsometry: Polarisation

Data Source

PatentUS10340196B1Method and system for selection of metrology targets for use in focus and dose applications
Publication Date: 2019.07.02 KLA CORP
  • US10340196B1 patent drawing
  • US10340196B1 patent drawing
  • US10340196B1 patent drawing

AI summary

The selection of metrology targets for use in a focus and dose application includes providing a FEM wafer including a plurality of fields with one or more metrology targets, measuring the one or more metrology targets within each field of the FEM wafer, performing a regression process on measurement results from the one or more selected fields of the FEM wafer to determine one or more DOI values for the one or more metrology targets of the one or more selected fields, calculating one or more diagnostic parameters for the one or more metrology targets of the one or more selected fields based on the regression process performed on the one or more selected fields of the FEM wafer, and identifying a set of candidate metrology targets based on the one or more calculated diagnostic parameters of the one or more selected fields of the FEM wafer.