Semiconductor Defect Detection via Interferometric Topography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical inspection tools, such as bright-field and dark-field detection tools, struggle to detect smaller defects on shrinking semiconductor devices due to their limitations in resolving smaller features and complex designs.
Innovation Solution
The use of surface topography data from interferometers to detect defects by comparing the topography characteristics of a test region on a sample with those of a reference surface, identifying variations that indicate defects through size, height, shape, and texture analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional bright-field and dark-field optical inspection tools are used, then the inspection process is simple and fast, but the ability to detect smaller defects on shrinking semiconductor devices deteriorates
Solution Approach 1:
The patent transitions from 2D optical reflection detection to 3D surface topography measurement. By using interferometry to capture height information (z-axis) in addition to x-y position, the system gains a new dimensional parameter for defect detection. This allows differentiation between actual defects and optical artifacts based on their three-dimensional characteristics, significantly improving detection precision for sub-resolution defects.
Solution Approach 2:
The patent replaces conventional optical reflection-based detection with interferometric topography measurement. Instead of relying on light scattering from defects, the system uses interference patterns to directly measure surface height variations. This substitution enables detection of defects that are smaller than the optical resolution limit, as it measures the physical topography rather than relying on optical contrast.
2Manufacturing precision
If the size of semiconductor devices continues to shrink, then device density and integration increase, but the difficulty of detecting smaller defects increases
Solution Approach 1:
The patent replaces conventional optical reflection-based detection with interferometric topography measurement. Instead of relying on light scattering from defects, the system uses interference patterns to directly measure surface height variations. This substitution enables detection of defects that are smaller than the optical resolution limit, as it measures the physical topography rather than relying on optical contrast.
Solution Approach 2:
The patent changes the measurement parameter from optical reflection intensity to surface height (topography). By measuring the z-height of surface features rather than their optical brightness, the system can detect defects based on their physical presence and shape, independent of their optical contrast. This parameter change allows detection of defects much smaller than the wavelength of light used.
3Productivity
If conventional optical inspection is used, then inspection speed is maintained, but the ability to detect defects on complex designs deteriorates
Solution Approach 1:
The patent replaces conventional optical reflection-based detection with interferometric topography measurement. Instead of relying on light scattering from defects, the system uses interference patterns to directly measure surface height variations. This substitution enables detection of defects that are smaller than the optical resolution limit, as it measures the physical topography rather than relying on optical contrast.
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 the rapid and accurate detection of defects like residue, dishing, edge roll-off, and hotspots on semiconductor wafers, improving yield and reducing downstream processing issues by communicating defect data to adjust fabrication processes.
Implementation Method 1
obtaining optical data from the sample using an interferometer, the optical data comprising a plurality of pixels
Data Source
AI summary
Defects are detected using surface topography data. The defects may be detected by determining topography characteristics within a region of interest on a sample, and the same topography characteristics of at least one reference surface. By comparing the topography characteristics in the region of interest for the sample and reference surface, common pattern structures may be removed, leaving only variations, which may be used to identify the presence of defects. For example, thresholds may be used to identify variations in the topography characteristics as defect candidates. Defects may be identified based on, e.g., size, height, shape, texture, etc. of candidate defects. In some implementations, rather than using a reference surface, the topography characteristic of the surface within the region of interest may be inspected based on prior knowledge of a required surface topography for the region of interest to determine if a defect is present.