Multi-scale defect classification for IC fabrication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing defect binning methods in integrated circuit fabrication fail to accurately classify macro-defects, leading to misclassification of micro-defects and contamination of defect bins, resulting in low-value inspection results and high false alarm rates due to incomplete information about macro-defects.
Innovation Solution
A computerized method that analyzes spatial clusters of defects to identify and reclassify micro-defects associated with macro-defects, using a processor to change micro-defect codes to macro-defect codes, thereby separating macro-defects from micro-defect bins for un-confounded analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If inspection tools detect defects larger than their designed feature size, then macro-defects can be detected, but they are reported as multiple smaller defects instead of a single defect
Solution Approach 1:
The patent segments the defect analysis process into two distinct levels: individual defect classification (micro-defects) and spatial cluster analysis (macro-defects). By dividing the defect population into clustered and non-clustered groups, the system preserves macro-defect information while maintaining accurate micro-defect classification.
Solution Approach 2:
The patent transitions from analyzing defects only at the individual level to incorporating spatial dimensionality by analyzing clusters of defects. This dimensional expansion allows the system to recognize macro-defects as collections of micro-defects while maintaining the ability to classify individual defect types.
2Device complexity
If existing defect binning methods classify each detected defect as an individual entity, then classification simplicity is maintained, but macro-defects are missed and defect bins are polluted
Solution Approach 1:
The classification process is segmented into two independent stages: first, individual defect classification using existing binning methods; second, spatial cluster analysis to identify macro-defects. This segmentation maintains the simplicity of individual defect classification while adding macro-defect recognition capability.
Solution Approach 2:
The patent performs preliminary classification of individual defects before conducting spatial cluster analysis. This preliminary action ensures that each defect is classified according to its inherent characteristics, and then clusters are identified based on spatial relationships, preventing bin pollution while maintaining classification accuracy.
3Productivity
If micro-defect classification is performed using local properties of reported defects, then classification speed is maintained, but statistical uncertainty increases and false alarms occur
Solution Approach 1:
The patent segments the defect population into clustered and non-clustered groups, applying different analysis methods to each segment. Non-clustered defects are classified using fast local property methods, while clustered defects are analyzed separately to determine macro-defect status, optimizing both speed and accuracy.
Solution Approach 2:
The patent applies local property classification only to non-clustered defects, while applying additional spatial analysis to clustered defects. This partial application of classification methods reduces computational overhead for the majority of defects while ensuring accurate classification of problematic clustered defects.
Data Source
AI summary
A computerized method for categorizing defects on a substrate. A list of defects on the substrate is received as input to a processor, where each defect is represented by a defect location and an associated micro-defect code. The input is analyzed with the processor to detect spatial clusters of defects on the substrate. The spatial clusters are analyzed with the processor to determine which of the spatial clusters represent known macro-defects and which of the spatial clusters represent unknown macro-defects. The micro-defect code associated with each defect that is included in one of the spatial clusters that is determined to be a known macro-defect is changed with the processor with a macro-defect code that is associated solely with the known macro-defect. The processor analyzes the defects that are included in one of the spatial clusters that is determined to be an unknown macro-defect to determine a predominantly occurring micro-defect code. The processor changes the micro-defect code associated with each defect that is included in the one spatial cluster that is determined to be an unknown macro-defect with the predominantly-occurring micro-defect code. The processor sends the changed list of defects on the substrate as output.