Wafer Surface Inspection via 3D Light Scattering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional surface inspection methods for semiconductor wafers are inefficient due to point-by-point low-speed scanning and uniform threshold settings, which restrict detection sensitivity and increase measurement time, especially when dealing with non-uniform surface roughness and crystal orientations.
Innovation Solution
The use of high-speed sampling devices to acquire three-dimensional data, including detection intervals and frequency, for creating region-specific maps and setting region-specific threshold levels, enabling three-dimensional simultaneous defect determination with spatially independent detectors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If point-by-point scanning with high-resolution sensor is used, then measurement precision is improved, but productivity deteriorates due to low speed and long measurement time
Solution Approach 1:
The patent divides the measurement task into two segments: a small-area high-resolution measurement region measured with an atomic force microscope for precision, and a large-area low-resolution measurement region measured with a light scattering apparatus for speed. This segmentation allows each method to be applied where it is most effective, resolving the contradiction between precision and productivity.
Solution Approach 2:
The patent transitions from a single-point measurement approach to a multi-dimensional measurement strategy by combining measurements from different spatial scales and different physical domains (mechanical contact vs. optical scattering). This dimensional approach enables simultaneous achievement of high precision in critical areas and high productivity in overall coverage.
2Ease of operation
If uniform threshold levels are set for entire surface, then ease of operation is improved, but measurement precision deteriorates in regions with different noise levels
Solution Approach 1:
The patent applies local quality by setting different threshold levels for different measurement regions based on their specific characteristics. The determination apparatus calculates separate threshold values for the high-resolution measurement region and the low-resolution measurement region, allowing each region to have optimized detection parameters matched to its noise level and measurement quality, thereby improving defect detection sensitivity while maintaining operational simplicity through automated regional differentiation.
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 allows for more accurate and efficient surface state analysis, significantly reducing measurement time, enhancing defect inspection sensitivity, and improving the detection of orientation features on semiconductor wafers.
Implementation Method 1
the signal processing of inspection object-scattered light that is one of the methods of inspecting the foreign substances and defects existing on a wafer has been based only upon the intensity of the light
Data Source
AI summary
Surface states have traditionally been measured with apparatuses such as an atomic force microscope (AFM), and these measurements have been high in resolution but low in speed. In conventional apparatuses for inspecting the foreign matter sticking to a wafer surface, and for inspecting defects present on the wafer surface, the inspection has had a tendency to be restricted in a region of the highest noise level arising from the roughness of the surface, the surface state, and/or crystal orientations, and thereby to reduce detection sensitivity in a region of lower noise levels. In these conventional techniques, signal processing of the light scattered from the object to be inspected has been based only upon the intensity of the light.This invention acquires three-dimensional data by, during such signal processing, adding detection intervals and the frequency of detection, as well as the intensity of light. The invention measures surface roughness of a target object (wafer) by creating region-specific three-dimensional maps from the three-dimensional data, then estimating the surface state of the wafer from analytical results, processing this estimated surface state as physical quantities, and analyzing the data.