Semiconductor Inspection Using Polarization Modulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current inspection apparatuses struggle to accurately distinguish between defects that affect semiconductor pattern quality, such as short-circuit and open-circuit defects, and edge roughness, especially when these defects are below the resolution limit of the optical system, leading to throughput degradation and inefficient defect detection.
Innovation Solution
An inspection apparatus utilizing a light source with a predetermined wavelength, a half-wave plate, and a splitter unit to rotate the polarization direction of light, allowing for the separation of defects through different polarization directions, and an image processor to calculate a polarization characteristic signal, which adjusts the polarization state to minimize the impact of edge roughness and enhance the detection of short-circuit and open-circuit defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a light source with wavelength below the pattern dimension is used to resolve fine patterns, then measurement precision is improved, but device complexity and cost increase significantly
Solution Approach 1:
The patent changes the parameter of light wavelength to match the pattern pitch (λ≥pattern pitch) rather than using shorter wavelengths, and combines this with polarization modulation to achieve defect detection without requiring ultra-high resolution optics, thus reducing system complexity while maintaining detection capability
2Measurement precision
If polarization modulation is applied to distinguish defect types, then measurement precision is improved, but device complexity increases due to additional optical components
Solution Approach 1:
The patent modulates the polarization state parameter of the illumination light and uses the polarization characteristics of scattered light from different defect types to achieve discrimination, enabling precise defect identification through parameter modulation rather than complex imaging systems
3Measurement precision
If edge roughness is filtered out to improve defect detection accuracy, then measurement precision is improved, but information loss occurs regarding actual pattern defects
Solution Approach 1:
The patent applies different evaluation criteria to different spatial frequencies: it suppresses low spatial frequency components corresponding to edge roughness while preserving and emphasizing high spatial frequency components corresponding to actual defects, thus achieving local quality differentiation in the signal processing
Solution Approach 2:
The patent changes the polarization state parameter and analyzes the polarization characteristics of scattered light, using the difference in polarization responses between edge roughness and actual defects to selectively filter information while preserving critical defect data
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 solution enables accurate detection of defects that impact semiconductor pattern quality without throughput degradation, effectively distinguishing between critical defects and non-critical edge roughness, even when they are below the optical system's resolution limit.
Implementation Method 1
a half-wave plate, and a splitter unit to rotate the polarization direction of light
Implementation Method 2
an illumination optical system that includes a light source emitting light having a predetermined wavelength, and illuminates a sample
Data Source
AI summary
An inspection apparatus comprising, an optical system emitting light having a predetermined wavelength, illuminating a sample while the light is converted into light having a polarization plane not in the range of −5 degrees to 5 degrees and 85 degrees to 95 degrees with respect to a direction of a repetitive pattern on the sample, an optical system for acquiring an image and forming said image on an image sensor using a lens, a half-wave plate, a first image sensor, a second image sensor, an inspection analyzer, wherein these differ in a transmission axis direction, a processor that obtains an average gray level and a standard deviation in each predetermined unit region of the image, and a defect detector, wherein a resolution limit defined by a wavelength of the light source and a numerical aperture of the lens is a value in which the pattern is not resolved.