Defect Classification via Multi-Angle Scattering Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current semiconductor inspection systems face challenges in detecting and classifying small defects on substrates due to limitations in sensitivity and throughput, particularly with methods like Energy-Dispersive X-Ray Spectrometry and Raman spectroscopy, which struggle with small particles and require long signal integration times.

Innovation Solution

The development of an inspection system that utilizes a narrowband illumination source to measure defect scattering characteristics such as scattering phase, intensity, and absorption, employing phase shifting phase contrast imaging and multiple detection modes to classify defects based on their material composition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional inspection methods (Energy-Dispersive X-Ray Spectrometry, Raman spectroscopy) are used to detect small defects, then material classification capability is provided, but sensitivity and throughput deteriorate due to long signal integration times

Engineering Contradiction:
Improvedefect detection sensitivityVSAvoidinspection throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent changes the detection parameter from conventional spectroscopic methods to scattering angle measurement. By measuring the angular distribution of scattered light at multiple angles simultaneously, the system achieves both high sensitivity for small particles and high throughput without requiring long integration times.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces a new measurement dimension by measuring scattering at multiple angles simultaneously rather than relying on single-angle spectroscopy. This multi-angle approach enables parallel extraction of material composition information, thereby improving throughput while maintaining sensitivity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If conventional inspection methods are used, then material classification is achieved, but detection capability for small particles deteriorates

Engineering Contradiction:
Improvematerial classification capabilityVSAvoidsmall particle detection capability
Core Design Contradiction:
Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent changes from spectroscopic parameter measurement to scattering angle parameter measurement. The angular distribution of scattered light provides material-specific signatures that enable both small particle detection and material classification simultaneously.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent segments the scattering pattern into multiple angular components, with each angle providing independent material information. This segmentation allows simultaneous detection of small particles and classification of material composition through multi-angle analysis.

Inventive Principle:
Principle #1Segmentation

3Device complexity

If single detection mode is used, then system simplicity is maintained, but defect classification accuracy deteriorates

Engineering Contradiction:
Improvesystem complexityVSAvoiddefect classification accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent adds the dimension of scattering angle to the detection system. By measuring at multiple angles simultaneously, the system gains additional classification information without requiring complex sequential measurement procedures, thus improving accuracy while keeping the system relatively simple.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

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 enables sensitive and high-throughput detection and classification of defects, including small particles, by efficiently utilizing spectral energy and providing accurate material identification suitable for semiconductor fabrication environments.

Implementation Method 1

the radiation emanating from the sample includes radiation specularly reflected by the sample and radiation scattered by the sample

Methodology Applied
Scientific EffectScattering: Scattering

Implementation Method 2

radiation specularly reflected by the sample

Methodology Applied
Scientific EffectSpecular reflection: Reflection

Implementation Method 3

a phase control device to introduce two or more different selected phase offsets between the radiation specularly reflected by the sample and the radiation scattered by the sample

Methodology Applied
Scientific EffectPhase shifting: Phase Modulation

Data Source

PatentUS10670537B2Systems and methods for defect material classification
Publication Date: 2020.06.02 KLA CORP
  • US10670537B2 patent drawing
  • US10670537B2 patent drawing
  • US10670537B2 patent drawing

AI summary

A inspection system includes an illumination source to generate an illumination beam, focusing elements to direct the illumination beam to a sample, a detector, collection elements configured to direct radiation emanating from the sample to the detector, a detection mode control device to image the sample in two or more detection modes such that the detector generates two or more collection signals based on the two or more detection modes, and a controller. Radiation emanating from the sample includes at least radiation specularly reflected by the sample and radiation scattered by the sample. The controller determines defect scattering characteristics associated with radiation scattered by defects on the sample based on the two or more collection signals. The controller also classifies the one or more particles according to a set of predetermined defect classifications based on the one or more defect scattering characteristics.