Optical Metrology Speckle Analysis for Roughness Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring roughness in semiconductor devices, such as SEM, TEM, and AFM, face limitations in throughput and flexibility, while optical metrology techniques rely on artificial assumptions and suffer from low sensitivity, especially in characterizing complex structures and random roughness.
Innovation Solution
The development of high-throughput optical metrology systems using beam profile reflectrometry with high spatial coherence illumination, which generates a speckle field by interfering specular and diffusely scattered light, allowing for the determination of residual signals indicative of structural irregularities through analysis of interference patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional roughness measurement techniques (SEM, TEM, AFM) are used, then measurement precision is achieved, but productivity is reduced due to low throughput and potential sample destruction
Solution Approach 1:
The patent replaces mechanical measurement systems (SEM, TEM, AFM) with an optical metrology system that uses light scattering and speckle analysis to measure roughness. This substitution enables non-contact, non-destructive measurement with high throughput capability, resolving the contradiction between measurement precision and productivity
Solution Approach 2:
The patent changes the measurement parameter from direct surface topography (mechanical/electronic) to optical scattering characteristics. By analyzing speckle patterns and light scattering behavior, the system achieves roughness measurement with both high precision and high throughput, eliminating the need for slow mechanical scanning or destructive cross-sectioning
2Productivity
If optical metrology techniques are used, then productivity is improved with high throughput, but measurement precision deteriorates due to reliance on artificial assumptions and low sensitivity
Solution Approach 1:
The patent introduces speckle patterns as an intermediary that carries information about surface roughness. By analyzing the statistical properties and spatial characteristics of speckle patterns, the system achieves high-sensitivity roughness measurement while maintaining high throughput optical measurement capabilities
Solution Approach 2:
The patent utilizes changes in optical scattering characteristics (analogous to color changes) to detect roughness. By measuring variations in light scattering intensity and speckle pattern statistics, the system achieves high measurement sensitivity without sacrificing throughput, eliminating the need for artificial assumptions about surface geometry
3Productivity
If scatterometry implementations are used, then productivity is improved, but measurement precision deteriorates due to inability to handle real roughness that causes light scattering outside of well-defined diffraction orders
Solution Approach 1:
The patent transitions from analyzing light intensity in diffraction orders (1D angular distribution) to analyzing speckle patterns in spatial domain (2D intensity distribution). This dimensional change enables detection of light scattered outside well-defined diffraction orders, achieving accurate real roughness characterization while maintaining high throughput optical measurement
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 enhances sensitivity and throughput by two orders of magnitude compared to pure dark-field metrology, effectively measuring structural irregularities and roughness parameters with increased accuracy and reliability.
Implementation Method 1
generates a speckle field by interfering specular and diffusely scattered light
Implementation Method 2
light scattered outside of the specularly reflected component of the diffracted light
Data Source
AI summary
Systems and methods are presented to enhance and isolate residual signals indicative of the speckle field based on measurements taken by optically based metrology systems. Structural irregularities such as roughness and topographical errors give rise to light scattered outside of the specularly reflected component of the diffracted light. The scattered light interferes constructively or destructively with the specular component in a high numerical aperture illumination and detection system to form a speckle field. Various methods of determining residual signals indicative of the speckle field are presented. Furthermore, various methods of determining structural irregularities based on analysis of the residual signals are presented. In various embodiments, illumination with a high degree of spatial coherence is provided over any of a wide range of angles of incidence, multiple polarization channels, and multiple wavelength channels. In addition, diffracted light is collected over a wide range of angles of detection.